Hey mr. Flynn, how do you analyze the the model? Just with deriving it and looking at rate of change and turning point, How do you get conclusions from that? Also, isnt the price of bitcoin wayyy too volatile to predict, for example the quintic function in this vid cant predict anything beyond the data. Im confused on how to phrase a RQ with this type of model- I want to do something like Why did ethereum crash? but this type of IA doesnt really answer that, right? Cause if i look at what happend in the day that it crashed, that isnt really maths is it? If I were to predict the the future price then all my ia would be in subbing in an x value to whatever function i get, with no analysis which isnt really the level of AA HL which im taking. But what im doing now is using newtonian interpolation (outside the scope od Math AA HL) to try and bring the level up? Do you think this is a good idea? I still dont quite understand how to analyse my data properly. Any help would be greatly appreciated!
Hi Vladimir. This is a lot to get through. The main point of the video was to show you how to use Geogebra to model data. Not necessarily to model bitcoin. You are correct that bitcoin is volatile and using your model to predict outside the data may not be a good idea. But all of this is good opportunity to reflect and show your understanding of modelling. Looking at what happened 'that day' isn't maths, but finding when that day was by finding key points is. I'm not convinced using Newtonian interpolation would be a good idea. If you really understand it and can explain it so a peer could understand it then maybe.
Sorry for the additional comment but what do the degrees of the polynomial function indicate? When the degree is higher (like 4, 5) it tends to represent all the point more. How can you use this for analysis?
Hi Mr Flynn! I am planning to model the price of ethereum and predict its future price. However, I have a few questions. First, should I use all the historical data (i.e. 2015 to 2022) or should I only use the data for one year to make the model? Second, regarding its future price, if the curve is similar to that of bitcoin in the video, how do I predict its price? Third, if I were to comment on the accuracy of the beft-fit line, should I use the R^2 value to determine it? Or should I comment based on the graph? Thank you!
1) This is up to you. You could do both. I think one year is better however. 2) To predict the future price, you can sub the future year (x value) into the function to get the future price (y value). 3) You should comment based on the graph but you can use the R^2 value as further evidence.
how did it turn out? Also, how do you analyze the the model? Just with deriving it and looking at rate of change and turning point, How do you get conclusions from that? Also, isnt the price of bitcoin wayyy too volatile to predict, for example the quintic function mr flynn uses can predict anything beyond the data he has
is modelling accurately and analyzing it enough for an IA of this sort? My idea is to get data, plot it, model it with a function, use calculus to analyze and draw conclusions but I feel like I need more as this is too simple. It's just applying derivatives a few times and seeing how that connects to real world events
@@incisuaslan8158 Hi there İnci, I modeled the price of bitcoin against the price of dogecoin and investigated why it fell during a particular period and explained the sudden spike in price in using another model
@@raissak.1875 hey raissa, what course do u take? was this an HL? also what do you mean by against? what was ur RQ and why did you have to explain why the price suddenly spiked? also what was ur other model lol any answer would be appreciated
Hi Mr. Flynn, I'm aiming to predict the price of bitcoin by the end of 2021. Could you suggest any methods/ topics from the math AI SL syllabus, that i can possibly incorporate along with the modelling aspect?
You could do a statistical analysis of the price over the last year or month or whatever. Find the mean, median, standard deviation etc. and explore how these might help you determine the future price.
Hi Mr. Flynn, thank you for your video, it's very helpful! I am planning to do something similar that also involves modeling data and calculus (looking at the relationship between the tax on a pack of cigarette in dollars (x values) and the consumption of cigarettes per capita (y values) to work out the most effective tax rate to deter consumption, but I'm a bit lost. I know that I need to first model the data, but how many models should I test for, and with GeoGebra, do I still work out each equation by hand (but it already gives me the full function) or just get straight into the analysis? Secondly, I believe I should calculate the rate of change to find the effective tax rate, but I'm not sure what the rate of change would represent, and do I calculate that for every data point? I know that that's a lot of specific questions, but I would really appreciate it if you can shed some light on this topic.
Hi Jen Jen, it depends what the data looks like. I would suggest plotting the data first before deciding whether to continue with this topic because I'm not sure it will work. Would it not just be the higher the tax rate, the lower the consumption? To find the most effective tax rate, you want to find where the consumption is at a minimum. The rate of change in your example would give you the change in the number of smokers for every dollar in tax (or whatever a unit of your x axis is). Once you have the derivative, you can calculate the rate of change for any point that is interesting to you and your IA. Hope that makes sense.
Depends what exactly your aim is. You could go into more detail about the functions and possibly calculate the R squared value by hand. You might want to find a more interesting function than a polynomial. You could try to include a statistical analysis along with the model but you run the risk of losing conciseness. Let me know if you have any ideas.
Hello Mr. Flynn your videos are awesome, thank you for all you do! I’m trying to model the pharmacokinetic profile of a medical drug for my Math AA SL IA. My data points suggest that nonlinear regression is appropriate to model the data. However I am not sure which method to use for modeling, do you think Automated nonlinear regression & using the hoerl function to model my data is good? Please help me out it will be much appreciated!!
I like the pharmacokinetic idea. I'm not familiar with the hoerl function. Have you watched my bitcoin or jumeirah beach hotel videos? They should help you with modelling if you have the data.
@@MrFlynnIB Thank you for the reply !! Will watch right now! Also i got some suggestions to model it and then investigate methods such as the trapezoidal rule and integration by parts to find the Area under the curve. Do you think these are appropriate?
Hi Mr. Flynn, amazing video for the examples, really help clear things out. I decided to also model using GeoGebra like you showed in the video, but I've been wondering if it's okay for us to directly take the equations made by the GeoGebra? Like in the video, is it okay if I directly take the polynomial equation and insert in my IA? And I directly go to discuss on the trends of the graph?
Hi Mr Flynn, I am planning to discover whether there may be a correlation between the optimal buying and selling period when stock prices are at their lowest and highest during a global crisis and the period within the global crises. Im not sure what types of mathematics would be beneficial to help my with my IA, I am maths SL AI so i can only use maths on my syllabus, so what would you suggest? Thank you!
You can use linear regression and correlation. Think what will be your x axis and what will be your y axis? I'm not entirely sure what you want your two variables to be.
Hello Mr Flynn, I have read on a comment that you prefer calculus in order to look at the rate of changes of price, instead of correlation. I was thinking of looking for the correlation between time and 3 cryptocurrencies (Ethereum, binance coin, and dogecoin) but now I am doubting about it. In your personal opinion what would you suggest to make pearson correlation and cryptocurrencies at the same time for my IA. Greetings, Daniel.
Hi Daniel. Correlation between cryptos is fine. I just mentioned that I found the rate of change more interesting especially for HL. What do you mean when you say correlation between time and 3 cryptos? Would it not be better to look at Eth v Bnb, then Eth v Doge then Doge v Bnb and compare. Or is that what you mean?
Hi Mr Flynn, firstly, you're a big help for IB Math students around, keep up the good work! Now, I want to do this for my Math IA, I'm in Math AA HL. Unfortunately for my lcassmates and I, we haven't gone too far in class in our sylllabus as of now. So what topics should I study in order to be able to do this IA well?
Did it turn out well? how do you analyze the the model? Just with deriving it and looking at rate of change and turning point, How do you get conclusions from that? Also, isnt the price of bitcoin wayyy too volatile to predict, for example the quintic function mr flynn uses can predict anything beyond the data he has Im confused on how to phrase a RQ with this type of model- something like Why did ethereum crash? but this type of IA doesnt really answer that, right?
Hi Mr. Flynn, thank you for your video! It's really great. I have a question, if it's a good idea to write about modelling the price of Ethereum from 2021 to 2022 with the possible impact of the war or even without that aspect? Sending kind regards!
Hello Mr Flynn, you mentioned something about a sinusoidal curve. Can it qualify to be one if all the points do not fit the curve. And if someone wants to use the sinusoidal curve to analyze water levels, what should they do?
Yes. A model doesn't mean that it has to fit perfectly. You can also use a non perfect model as an opportunity to critically reflect. Sinusoidal model for water levels could be a good IA. Watch my video on modelling the Dubai Eye. This will help.
how did it turn out? Also, how do you analyze the the model? Just with deriving it and looking at rate of change and turning point, How do you get conclusions from that? Also, isnt the price of bitcoin wayyy too volatile to predict, for example the quintic function mr flynn uses can predict anything beyond the data he has
Hello Mr Flynn, my math teacher says i can automatically get a function for my graph without trying out any polynomials, but he hasnt told me how i can do that and i have tried everything but failed, would you know how i could do that?
Do you mean "can't automatically...". You can try applying transformations to a polynomial until it looks like it fits. You could also sub different points into a general polynomial and solve simultaneous equations. Finally you could do polynomial regression which is essentially what the computer does. This is very complicated by hand and I wouldn't recommend it. Even for linear regression it is complicated. None of those options work great which is why I recommend letting Geogebra do it.
Hey Mr. Flynn for my maths hl ia I was planning to study the correlation between the responsiveness of Elon’s musks tweets and Bitcoin price. I was thinking of adding the rate of responsiveness as well to further complicate things and I wanted to hear what you think of this idea. Would this be a 7 ia.
Hi Zahi. How are you going to quantify Elon's tweets? Are you going to look at number of comments or likes? Ask yourelf 'What is the x axis and what is the y axis?'. It could work but I don't think it would be a straight forward correlation. I think it would be better to model the price like I have done and then find the rate of change of price just after he tweets. Plenty to discuss there. Let me know what you think.
@@MrFlynnIB Thank you Mr Flynn modelling the price makes more sense, I will put more thought into it and probably create a rq which is focused to just about 5 tweets.
hey, if i was to do this and use all the data to predict what will happen in the next months what topics would i be focusing on? is it more of functions? or would it also incorporate number and algebra?
Hi Mario. Sorry I missed this comment. It depends on your aim and the direction you want to take it but the main idea that I suggest in the video focuses on functions (modelling) and calculus.
Hey, Mr. Flinn I’m going to do my IA about the predicted price of a single Tesla’s stock for the next 5 months and I was wondering what mathmatical methods I can use for this. I can for sure make a function and find the best fit according to the past trend but wouldn’t it be too simple? It would be really helpful if you can help me come up with maths I can use!
Hi Nathan. Ya sure you can do this. It is similar to my bitcoin idea. You can use modelling and then calculus to analyze rates of change. Did you watch the video?
Hey Mr. Flynn your video is great! I'm currently study math AI HL and I'm planning on doing a IA to predict the future trend of Johnson & Johnson's stock return with the best fit polynomial line that you referred to in the vid. However, my teacher suggested that I need another model or method to predict the future trend so that I can compare the prediction result from the method to the result I have from the best fit polynomial model. Could you please suggest any other method that you might find it suitable? Thank you so much!
Not sure. Maybe you could look at what experts are predicting. You could also maybe assume that yearly growth stays the same. What I'm doing in my video is looking at the past to predict the future. So obviously it might not be perfectly accurate but you can talk about the assumptions you are making.
Can I know by the equation that you calculated which currency Bitcoin and others or Euro dollars and Pound dollars and which currency I know where the price goes up or down
Hello Mr.Flynn, great video on the IA examples. After going through all of the options, I chose this topic as my IA topic and im trying to do something similar by modelling and comparing 2 separate crypto currencies and by doing this, pinpoint the price hikes, price drops. I'm unsure of how to find the derivative of the btc graph to find the maximum and the minimum point. Some help on this would be much appreciated 😁
Hi Mr. Flynn, I'm thinking of modelling the popularity of NFTs based on how many sales and using data from the last 2-3 years. I'm not sure what to do next so that my IA is deep enough and demonstrates enough skill. Could you give me some ideas? Thank you.
I wanted to say thanks and that I’ll look into finding an aim but I have another prospective idea. I have this floor lamp at home and I was wondering if I could model like a lamp shade or cover that disperses the light in a better way or one that would disperse the light equally. Or I could find the surface area or volume of the lamp head thing and compare it to lamps that have been proved to share the most light equally.
@@MrFlynnIB hello again sir, so i wanted to come up with an RQ for the idea with dispersion of light but there have not been a lot of studies on that sort of thing so i wouldn’t have a guide, also it’s a very physics heavy topic i think, and i don’t do physics. i was thinking of modeling the volume of revolution of the lamp shade like you did in a video of yours, i also have this weird vase that i could model the revolution of. do you think this works better, again thank you so much for your help
Hi Mr. Flynn! thank you so much for your video, it's very helpful it actually gave me an idea of what to do for my maths IA! I am planning to do something similar tho maybe modeling then comparing or finding the correlation, between Ethihad and emirates tickets prices from one destination to the other. I was wondering what question is better? I'm kinda a bit lost on what to do. if you could please give me an idea similar to the one i have i would really appreciate it!
I would imagine there is a strong correlation between ticket prices from one destination to another but this doesn't seem that interesting to me. Maybe correlation between distance of a flight and ticket price would be better. Then you can use your model to predict how much a flight should cost. Even better might be to model the price of a single flight over time to see how covid affected the price. That would be similar to my bitcoin idea but price of a flight instead of price of a bitcoin.
@@MrFlynnIB Thank you for your suggestion, So do you think I should model the price of a single flight over time (for both airlines) and on how covid affected the price. so then I can compare both airlines but what would I say if I compared them (does this makes sense? sorry) and thank u
or i got another idea too! maybe I could model the price of a single flight over time (so like I could gather data of the whole month of March and the whole month of august before covid and now) for just one airline and compare the dates maybe like to find the cheapest holiday to go to so it will be like spring break or summer?
Hey Mr. Flynn, you did a great job with this video, it definitely provided some really helpful suggestions. I was thinking of modelling different countries' military expenditure (ex. U.S., China, and Russia, etc.). Aside from the different ways of modelling that you talked about in the video, as well as the derivatives, is there anything else that I should consider? Any advice/suggestions would be greatly appreciated, and thank you once again for the video.
@@MrFlynnIB yes; and my aim was to examine how long it would take for a country such as China to spend just as much if not more than the U.S., as well as the total amount that would have been needed to do so.
@@MrFlynnIB Thank you very much for the feedback; do you also think I should derive the modelling equations and examine those for some extra complexity?
@@woreandpiece4815 I wouldn't recommend it. Depending on the function, it can get really difficult and then really difficult to explain coherently. Let the computer get them but be sure to explain why you chose them. For example what is it about the data points that made you choose a quadratic function?
Mr Flynn, I'm starting my IA shortly and I had a question. My original idea was investigating some tests of convergence for infinite series. I find the topic interesting and I would like to challenge myself into doing a topic outside of the curriculum. My other idea is something to do with modelling crypto (which led me here). What would you say would be my best option
Generally speaking (but not always), IAs that have a real- life application work better. If you can find a good real-life application for your infinite series, this could be very good. But you need a specific aim. 'Investigating different tests for convergence' isn't a good aim as it is too vague. I really like the modelling crypto idea (which is why I made a video on it). Let me know what you decide.
hello sir thank you for the amazing video, i was wondering if it was a feasible idea for me to do my IA on the popularity of a certain book in relation to viral tiktoks made about them. i would use the graph from google trends and find ways to make it more complex. this is for AA HL, it’s all i could think of to get high marks for personal engagement
Not sure that would work although you can collect the data first and see what it looks like. What is your x-axis and what is your y-axis? You are not thinking about personal engagement correctly. You don't have to have a personal interest in the topic to get a high score in the personal engagement criterion. It is more about how you engage with the exploration. For example, I don't have much interest in cricket but I could write an IA with high personal engagement. Conversely, I am very interested in football but I could write an IA with low personal engagement.
@@MrFlynnIB i see what you mean, i was just trying to find a topic that linked to novels in some way.the x axis would be time, the y axis would be the interest. metric on google trends. Alternatively I considered doing the volume of revolution of a cactus plant (a very small one) or a vase but i wasn’t sure what the aim would be. I’m still trying to find a topic that could link to books/reading in some way but i’ve had no luck
@@fade4992 It might work. Get the data and see. The cactus idea would be fine. The aim would just be to get the volume of the cactus. Why you would want to find it is the question.
@@MrFlynnIB lamp shade or cover that disperses the light in a better way or one that would disperse the light equally. Or I could find the surface area or volume of the lamp head thing and compare it to lamps that have been proved to share the most light equally.
Hi Mr. Flynn, i am an AI HL student and i was looking for topics for my IA. i am interested in this field and topic but i don't know what all math i will be using in this. Will using calculus to find various points and their significance be enough for an IA?
It depends what you do. Remember the aim is to use mathematics commensurate with the level of the course and to demonstrate understanding. More maths doesn't necessarily mean better. Having said that, you could look at bringing in some statistics to look at what the price might do. Lots of statistics in the AI HL course that you could use.
Hey Mr. Flynn. I am doing AI HL exploration on 'when will China's GDP cross USA's GDP', and my sir told me to use some calculus with modeling. So I wanted to know what calculus can I use for this and how?(my class hasn't reached calculus yet)
Calculus (differential calculus in particular) is all about rates of change. So you can use it to calculate the growth rate of the GDP. For example you could calculate the current growth rate and assume it stays constant to find when they will cross. I like the idea.
@@saumyashah152 It's possible but might get complicated. What I suggested in the previous comment actually involves solving a fairly simple differential equation.
hey Mr. Flynn I am an AA HL math student and I want to move forward with this Idea however i am struggling with finding math that is in the AA HL syllabus specifically math that is only for math aa hl to implement in this idea could you please help me out
A number of students have asked me similar questions about finding maths that is in the HL syllabus. Remember everything in the SL syllabus is also in the HL syllabus.
Hi Mr Flynn, I am wondering how maths from the HL syllabus could be applied to this? Because as far as I can see it's largely based on derivatives and SL calc. Thank you!
@@MrFlynnIB Thanks for the quick reply Mr Flynn. I know that all SL is within HL, but that is exactly the problem as I believe there is a need to refer to math that is out of SL but within HL? Thanks
@@daniccle You don't have to have maths outside the SL course but you need to demonstrate and better understanding and more sophistication than an SL student would.
Hey Mr Flynn!! Thanks for all the great video!! I am doing my math AI IA on modelling the force/impact required to make a knockout in martial arts. I'm kind of lost in terms of starting out with the whole thing. I'm panicking and I don't know what to do. It would be great if you could help me sir!!
@@MrFlynnIB omg thank you so much for replying. I just wanted to ask if you have any math suggestion I could do. My teacher wants a lot of maths involved, so. Regardless, thank you for helping me out!
@@yoongmoonsahennishuee7970 Lots of maths you could use. Modelling, calculus, solving equations. If your teacher wants more, you can look to add in some statistics.
Hi Mr. Flynn, thank you so much for this video. I was wondering if it'd be a good idea to model the prices of a certain type of cryptocurrency from late 2019 to late 2020 and analyze the impact COVID *possibly* had on the rise and fall of the prices?
@@aamirsangey5976 I talk about this in the video. By calculating the derivative, you can find the max and min points and the points of inflexion. You can also get growth rate at any point.
Hey Mr Flynn, I was just wondering if this would be somewhat suitable for a Math AI HL course, such as modelling the rise in the price of Ethereum. Do you think modelling in this form would be suitable for the HL course?
Ya sounds good. Maybe refine your aim to have a more specific goal. 'Model the price of Ethereum in order to predict its price on 1st of August 2021' is an example.
@@MrFlynnIB Could forms of statistical math also be used in this kind of IA, or would it deviate away from the calculus used in this topic? Thanks for the feedback
Hi Mr Flyyn, how are you ? I’m having a problem since When I write “Fit” in Geo Gebra no formulas appear to be selected as it occurred in your screen. Do you have any idea why this happens ? Regards !
How are you Mr Flynn? When modelling price of bitcoins, there is always the element of randomness when we predict future prices along with information from past pricing data. Thus, how do we incorporate this fact in our IA to make the predictions more accurate? Thank you.
Hi Xuanyi. Of course in reality it is very difficult to predict the price of bitcoin (or anything for that matter). Remember you can also talk about the assumptions you made and limitations of your model. This helps show understanding and reflection. What I'm trying to do in the video is use what has happened in the past to predict the future. You could say something like 'assuming that that the price behaves the same this month as last month, what will the price be?' and then talk about all the problems with that assumption. Let me know if that helps.
@@MrFlynnIB Hi Mr Flynn. This helps a lot. Greatly appreciated. One last question : how can I corporate calculus inside this modelling process? I see you mentioning this in the comment and talked about the point of inflexion. What are the purposes of these? Thanks :)
@@xuanyidu3259 So once you have the function, you can differentiate it to find the rate of growth at different points. You can also find any turning points. The point point of inflection is where the curvature changes. In this example it is where the rate of growth starts to decrease or increase which might be important in your model. I talk about this a bit in the video.
hi Mr Flynn I love your video. It really helped me make a final decision with my IA. I'm thinking of comparing the prices of ethereum and bitcoin over the years. But I can't seem to think of any math to include in my IA. I'm doing SL AI and so I'm very limited on this. i hope you can help :)
This sounds like a linear regression IA to me. Find the price of Ethereum and Bitcoin at particular times over a given period of time (e.g. 1st of every month for the last 5 years). Create a scatter plot and find the correlation and regression line. Then maybe make some predictions. If Bitcoin goes to 100k, what will Ethereum go to?
@@MrFlynnIB thank you for your feedback!! I'm actually doing exactly that, lol. I've decided to find the correlation between bitcoin and ethereum prices from may 2015 until may 2021, then finding the difference between 2015 and 2021 to make predictions for the next 6 years. I'm not sure if this makes sense yet.
@@Krisy444 Sounds good. Let me know how it works out. I believe the relationship won't be exactly linear. I think Ethereum has actually grown faster in the last 6 years. But that's fine. You can just mention that as you are reflecting.
Hello Mr Flynn I currently study HL AI math. If I do my IA on Modelling the price of Dogecoin in order to predict it's price in one year, how can I apply HL specific topics. Obviously, using calculus to look at the rate of change of price is a way or exponential regression. Can u please help out? How would I structure it out
Both of these ideas seem fine. Let your aim guide you. You could start with exponential regression and see where that takes you. I'd imagine the price of Dogecoin isn't going to grow exponentially for a full year so then you might have to adjust your model. Maybe a logistic model after the exponential one.
@@MrFlynnIB How can calculus be applied to this and is there any context to using calculus in this?, as for HL I aim to use calculus but my teacher said there needs to be context behind it, not just using it for the purpose of the IA. Can you help out with this?
@@danieldjerdjerian362 You can use calculus after you have created the model. Calculus will help you to find the rate of growth of the price of Dogecoin. You can also find the turning points and points of inflexion.
@@MrFlynnIB My teacher said I can do this too but was worried that this was only applicable for the needs of the IA and no bigger purpose, can you explain the context behind using calculus in this case?
@@danieldjerdjerian362 If you want to predict the price in the future, you need to understand how the price behaved in the past and rates of change can help you understand this.
Hi Mr Flynn, Thank you so much for the video! It was so useful. I'm thinking about modelling the popularity of rock music but I'm concerned if the topic is maths-y enough? Also, whether that's a strong enough topic for a maths IA for AA SL? Thanks :)
Sounds interesting. I don't see any reason why it wouldn't be a strong enough topic. Provided the popularity of rock music is changing. First get the data and have a look. If it moves and you can do something like I did with the bitcoin price then great.
Hi Mr Flyyn, thank you for all these great ideas ! I wanted to ask you, how could I develop them into words for my IA introduction ? , could I say something like: “In this investigation I will be modelling the price of Bitcoin, in order to find the function that best fits the model, and then analyse different points where the price of Bitcoin underwent significant alterations (such as a take off or a crash) to understand what events took place in those specific days.
I think you need to make your aim more concise (less words). Maybe 'my aim is to model the price of bitcoin in order to predict its value in 3 months' (or whatever time you want).
@@MrFlynnIB but just one thing, with which of the tools you mention in the video I could develop a possible prediction of the price ? Should I start testing until I find a function that best fits the past prices and seems to continue in a reasonable way ? (Meaning it doesn’t goe drastically down as you showed in the video). And lastly, how could derivatives be applied in the prediction of the price in for example 3 months ? Sorry for the long questions, and again thank you a lot !
@@tomasbalbal6958 you need to decide this for yourself. Maybe you could make a prediction on the assumption that the next three months will behave exactly like the last 3 months. Then you can find when the next dip in price will occur. Something like that.
Lessons on every topic, all IA videos and full past paper solutions at mrflynnib.com.
Hey mr. Flynn, how do you analyze the the model? Just with deriving it and looking at rate of change and turning point, How do you get conclusions from that? Also, isnt the price of bitcoin wayyy too volatile to predict, for example the quintic function in this vid cant predict anything beyond the data.
Im confused on how to phrase a RQ with this type of model- I want to do something like Why did ethereum crash? but this type of IA doesnt really answer that, right? Cause if i look at what happend in the day that it crashed, that isnt really maths is it? If I were to predict the the future price then all my ia would be in subbing in an x value to whatever function i get, with no analysis which isnt really the level of AA HL which im taking. But what im doing now is using newtonian interpolation (outside the scope od Math AA HL) to try and bring the level up? Do you think this is a good idea? I still dont quite understand how to analyse my data properly. Any help would be greatly appreciated!
Hi Vladimir. This is a lot to get through. The main point of the video was to show you how to use Geogebra to model data. Not necessarily to model bitcoin. You are correct that bitcoin is volatile and using your model to predict outside the data may not be a good idea. But all of this is good opportunity to reflect and show your understanding of modelling. Looking at what happened 'that day' isn't maths, but finding when that day was by finding key points is. I'm not convinced using Newtonian interpolation would be a good idea. If you really understand it and can explain it so a peer could understand it then maybe.
Hello Mr. Flynn. Is it possible to come up with a logical mathematical model which supposedly predicts the future prices through this process?
Sorry for the additional comment but what do the degrees of the polynomial function indicate? When the degree is higher (like 4, 5) it tends to represent all the point more. How can you use this for analysis?
Hi Mr Flynn! I am planning to model the price of ethereum and predict its future price. However, I have a few questions. First, should I use all the historical data (i.e. 2015 to 2022) or should I only use the data for one year to make the model? Second, regarding its future price, if the curve is similar to that of bitcoin in the video, how do I predict its price? Third, if I were to comment on the accuracy of the beft-fit line, should I use the R^2 value to determine it? Or should I comment based on the graph? Thank you!
1) This is up to you. You could do both. I think one year is better however.
2) To predict the future price, you can sub the future year (x value) into the function to get the future price (y value).
3) You should comment based on the graph but you can use the R^2 value as further evidence.
how did it turn out? Also, how do you analyze the the model? Just with deriving it and looking at rate of change and turning point, How do you get conclusions from that? Also, isnt the price of bitcoin wayyy too volatile to predict, for example the quintic function mr flynn uses can predict anything beyond the data he has
is modelling accurately and analyzing it enough for an IA of this sort? My idea is to get data, plot it, model it with a function, use calculus to analyze and draw conclusions but I feel like I need more as this is too simple. It's just applying derivatives a few times and seeing how that connects to real world events
Hi Mr Flynn, thank you so much for this video. With your help I managed to score a 6 in my math IA! Keep up the good work. Warm Regards :)
Great to hear. Well done.
Hi Raissa, what was your IA about?
@@incisuaslan8158 Hi there
İnci, I modeled the price of bitcoin against the price of dogecoin and investigated why it fell during a particular period and explained the sudden spike in price in using another model
@@raissak.1875 hey raissa, what course do u take? was this an HL? also what do you mean by against? what was ur RQ and why did you have to explain why the price suddenly spiked? also what was ur other model lol any answer would be appreciated
Hey Mr. Flynn, do you think this sort of math IA (modelling the price of a crypto) is appropriate for AA HL?
Ya sure. You probably want to go into more depth than an SL student would.
How did it turn out to be?
Hi Mr. Flynn, I'm aiming to predict the price of bitcoin by the end of 2021. Could you suggest any methods/ topics from the math AI SL syllabus, that i can possibly incorporate along with the modelling aspect?
You could do a statistical analysis of the price over the last year or month or whatever. Find the mean, median, standard deviation etc. and explore how these might help you determine the future price.
Hi Mr. Flynn, thank you for your video, it's very helpful! I am planning to do something similar that also involves modeling data and calculus (looking at the relationship between the tax on a pack of cigarette in dollars (x values) and the consumption of cigarettes per capita (y values) to work out the most effective tax rate to deter consumption, but I'm a bit lost.
I know that I need to first model the data, but how many models should I test for, and with GeoGebra, do I still work out each equation by hand (but it already gives me the full function) or just get straight into the analysis? Secondly, I believe I should calculate the rate of change to find the effective tax rate, but I'm not sure what the rate of change would represent, and do I calculate that for every data point? I know that that's a lot of specific questions, but I would really appreciate it if you can shed some light on this topic.
Hi Jen Jen, it depends what the data looks like. I would suggest plotting the data first before deciding whether to continue with this topic because I'm not sure it will work. Would it not just be the higher the tax rate, the lower the consumption? To find the most effective tax rate, you want to find where the consumption is at a minimum. The rate of change in your example would give you the change in the number of smokers for every dollar in tax (or whatever a unit of your x axis is). Once you have the derivative, you can calculate the rate of change for any point that is interesting to you and your IA. Hope that makes sense.
What further analysis would you suggest for HL?
Depends what exactly your aim is. You could go into more detail about the functions and possibly calculate the R squared value by hand. You might want to find a more interesting function than a polynomial. You could try to include a statistical analysis along with the model but you run the risk of losing conciseness. Let me know if you have any ideas.
@Mr. Flynn IB is this a good IA topic for Maths HL?
Hello Mr. Flynn your videos are awesome, thank you for all you do! I’m trying to model the pharmacokinetic profile of a medical drug for my Math AA SL IA. My data points suggest that nonlinear regression is appropriate to model the data. However I am not sure which method to use for modeling, do you think Automated nonlinear regression & using the hoerl function to model my data is good? Please help me out it will be much appreciated!!
I like the pharmacokinetic idea. I'm not familiar with the hoerl function. Have you watched my bitcoin or jumeirah beach hotel videos? They should help you with modelling if you have the data.
@@MrFlynnIB Thank you for the reply !! Will watch right now! Also i got some suggestions to model it and then investigate methods such as the trapezoidal rule and integration by parts to find the Area under the curve. Do you think these are appropriate?
@@frisf8258 You can use the trapezoidal rule if you want. You can use integration by parts but only if appropriate. Don't use it if you don't need to.
Thank you so much for these videos! You're saving my grade!
@MrFlynnIB can we do this for AAHL IA as well- or are there better ideas for an econs related one
Sure. Watch my two 'ideas' videos. Part 7 and 8 of the IA guide.
Hi Mr. Flynn, amazing video for the examples, really help clear things out. I decided to also model using GeoGebra like you showed in the video, but I've been wondering if it's okay for us to directly take the equations made by the GeoGebra? Like in the video, is it okay if I directly take the polynomial equation and insert in my IA? And I directly go to discuss on the trends of the graph?
Sure but make sure you explain where it comes from and why you chose that polynomial.
@@MrFlynnIB Ah i see, thank you so much!
Hi Mr Flynn, I am planning to discover whether there may be a correlation between the optimal buying and selling period when stock prices are at their lowest and highest during a global crisis and the period within the global crises. Im not sure what types of mathematics would be beneficial to help my with my IA, I am maths SL AI so i can only use maths on my syllabus, so what would you suggest? Thank you!
You can use linear regression and correlation. Think what will be your x axis and what will be your y axis? I'm not entirely sure what you want your two variables to be.
@@MrFlynnIB my two variables are price and time(days). Other than prive what il could the other variable be? Thank you!
Hello Mr Flynn, I have read on a comment that you prefer calculus in order to look at the rate of changes of price, instead of correlation. I was thinking of looking for the correlation between time and 3 cryptocurrencies (Ethereum, binance coin, and dogecoin) but now I am doubting about it. In your personal opinion what would you suggest to make pearson correlation and cryptocurrencies at the same time for my IA. Greetings, Daniel.
Hi Daniel. Correlation between cryptos is fine. I just mentioned that I found the rate of change more interesting especially for HL. What do you mean when you say correlation between time and 3 cryptos? Would it not be better to look at Eth v Bnb, then Eth v Doge then Doge v Bnb and compare. Or is that what you mean?
Hi Mr Flynn, firstly, you're a big help for IB Math students around, keep up the good work! Now, I want to do this for my Math IA, I'm in Math AA HL. Unfortunately for my lcassmates and I, we haven't gone too far in class in our sylllabus as of now. So what topics should I study in order to be able to do this IA well?
Depends on exactly what you want to do but I would say functions (modelling in particular) and differential calculus.
Did it turn out well? how do you analyze the the model? Just with deriving it and looking at rate of change and turning point, How do you get conclusions from that? Also, isnt the price of bitcoin wayyy too volatile to predict, for example the quintic function mr flynn uses can predict anything beyond the data he has
Im confused on how to phrase a RQ with this type of model- something like Why did ethereum crash? but this type of IA doesnt really answer that, right?
Hi Mr. Flynn, thank you for your video! It's really great. I have a question, if it's a good idea to write about modelling the price of Ethereum from 2021 to 2022 with the possible impact of the war or even without that aspect? Sending kind regards!
Ya sure. Similar to my bitcoin idea.
Hello Mr Flynn, you mentioned something about a sinusoidal curve. Can it qualify to be one if all the points do not fit the curve. And if someone wants to use the sinusoidal curve to analyze water levels, what should they do?
Yes. A model doesn't mean that it has to fit perfectly. You can also use a non perfect model as an opportunity to critically reflect. Sinusoidal model for water levels could be a good IA. Watch my video on modelling the Dubai Eye. This will help.
@@MrFlynnIB what do you mean by non-perfect model?
@@kaleebucynthia663 Like what you said. Where all the points do not fit the curve.
@@MrFlynnIB oh okay, thank you 🙏🏾
Do you do private tutoring on for the math IA?
Yes I do private tutoring. Feel free to email me at contact@mrflynnib.com.
Hi Mr. Flynn! Would Modelling and analyzing the price of bitcoin to predict future trends be a good IA topic for HL?
Ya sure.
how did it turn out? Also, how do you analyze the the model? Just with deriving it and looking at rate of change and turning point, How do you get conclusions from that? Also, isnt the price of bitcoin wayyy too volatile to predict, for example the quintic function mr flynn uses can predict anything beyond the data he has
@@vladimirputin7072 I didn't end up doing the topic
Hello Mr Flynn, my math teacher says i can automatically get a function for my graph without trying out any polynomials, but he hasnt told me how i can do that and i have tried everything but failed, would you know how i could do that?
Do you mean "can't automatically...". You can try applying transformations to a polynomial until it looks like it fits. You could also sub different points into a general polynomial and solve simultaneous equations. Finally you could do polynomial regression which is essentially what the computer does. This is very complicated by hand and I wouldn't recommend it. Even for linear regression it is complicated. None of those options work great which is why I recommend letting Geogebra do it.
Hey Mr. Flynn for my maths hl ia I was planning to study the correlation between the responsiveness of Elon’s musks tweets and Bitcoin price. I was thinking of adding the rate of responsiveness as well to further complicate things and I wanted to hear what you think of this idea. Would this be a 7 ia.
Hi Zahi. How are you going to quantify Elon's tweets? Are you going to look at number of comments or likes? Ask yourelf 'What is the x axis and what is the y axis?'. It could work but I don't think it would be a straight forward correlation. I think it would be better to model the price like I have done and then find the rate of change of price just after he tweets. Plenty to discuss there. Let me know what you think.
@@MrFlynnIB Thank you Mr Flynn modelling the price makes more sense, I will put more thought into it and probably create a rq which is focused to just about 5 tweets.
Thank you Mr.Flynn
hey, if i was to do this and use all the data to predict what will happen in the next months what topics would i be focusing on? is it more of functions? or would it also incorporate number and algebra?
Hi Mario. Sorry I missed this comment. It depends on your aim and the direction you want to take it but the main idea that I suggest in the video focuses on functions (modelling) and calculus.
Hey, Mr. Flinn I’m going to do my IA about the predicted price of a single Tesla’s stock for the next 5 months and I was wondering what mathmatical methods I can use for this. I can for sure make a function and find the best fit according to the past trend but wouldn’t it be too simple? It would be really helpful if you can help me come up with maths I can use!
Hi Nathan. Ya sure you can do this. It is similar to my bitcoin idea. You can use modelling and then calculus to analyze rates of change. Did you watch the video?
Hey Mr. Flynn your video is great! I'm currently study math AI HL and I'm planning on doing a IA to predict the future trend of Johnson & Johnson's stock return with the best fit polynomial line that you referred to in the vid. However, my teacher suggested that I need another model or method to predict the future trend so that I can compare the prediction result from the method to the result I have from the best fit polynomial model. Could you please suggest any other method that you might find it suitable? Thank you so much!
Not sure. Maybe you could look at what experts are predicting. You could also maybe assume that yearly growth stays the same. What I'm doing in my video is looking at the past to predict the future. So obviously it might not be perfectly accurate but you can talk about the assumptions you are making.
Can I know by the equation that you calculated which currency Bitcoin and others or Euro dollars and Pound dollars and which currency I know where the price goes up or down
Ya if the function is increasing, the price is going up. If the function is decreasing, the price is going down.
Hello Mr.Flynn, great video on the IA examples. After going through all of the options, I chose this topic as my IA topic and im trying to do something similar by modelling and comparing 2 separate crypto currencies and by doing this, pinpoint the price hikes, price drops. I'm unsure of how to find the derivative of the btc graph to find the maximum and the minimum point. Some help on this would be much appreciated 😁
What did you end up doing ; I'm having trouble with finding maximum minimum points etc.
Hi Mr. Flynn,
I'm thinking of modelling the popularity of NFTs based on how many sales and using data from the last 2-3 years. I'm not sure what to do next so that my IA is deep enough and demonstrates enough skill. Could you give me some ideas? Thank you.
Ya sounds good. You can look at trends, perform some calculus on the function. Maybe some statistics if needed. Try and have a precise aim.
I wanted to say thanks and that I’ll look into finding an aim but I have another prospective idea. I have this floor lamp at home and I was wondering if I could model like a
lamp shade or cover that disperses the light in a better way or one that would disperse the light equally. Or I could find the surface area or volume of the lamp head thing and compare it to lamps that have been proved to share the most light equally.
Also are we not allowed to send links in the comments?
My comment kept getting deleted when I sent the link to the lamp
@@fade4992 I like the lamp shade idea. It sounds interesting. Ya I thought you could post links in the comments. I have done it before.
@@MrFlynnIB Thank you so much!
@@MrFlynnIB
hello again sir, so i wanted to come up with an RQ for the idea with dispersion of light but there have not been a lot of studies on that sort of thing so i wouldn’t have a guide, also it’s a very physics heavy topic i think, and i don’t do physics. i was thinking of modeling the volume of revolution of the lamp shade like you did in a video of yours, i also have this weird vase that i could model the revolution of. do you think this works better, again thank you so much for your help
@@fade4992 probably easier although not as interesting. Remember to consider why are you finding the volume.
Hi Mr. Flynn! thank you so much for your video, it's very helpful it actually gave me an idea of what to do for my maths IA! I am planning to do something similar tho maybe modeling then comparing or finding the correlation, between Ethihad and emirates tickets prices from one destination to the other. I was wondering what question is better? I'm kinda a bit lost on what to do. if you could please give me an idea similar to the one i have i would really appreciate it!
I would imagine there is a strong correlation between ticket prices from one destination to another but this doesn't seem that interesting to me. Maybe correlation between distance of a flight and ticket price would be better. Then you can use your model to predict how much a flight should cost. Even better might be to model the price of a single flight over time to see how covid affected the price. That would be similar to my bitcoin idea but price of a flight instead of price of a bitcoin.
@@MrFlynnIB Thank you for your suggestion, So do you think I should model the price of a single flight over time (for both airlines) and on how covid affected the price. so then I can compare both airlines but what would I say if I compared them (does this makes sense? sorry) and thank u
or i got another idea too! maybe I could model the price of a single flight over time (so like I could gather data of the whole month of March and the whole month of august before covid and now) for just one airline and compare the dates maybe like to find the cheapest holiday to go to so it will be like spring break or summer?
@@shamasdiary2728 Ya perfect. You have a nice clear aim then too.
@@MrFlynnIB Thank you so muchh
Hey Mr. Flynn, you did a great job with this video, it definitely provided some really helpful suggestions. I was thinking of modelling different countries' military expenditure (ex. U.S., China, and Russia, etc.).
Aside from the different ways of modelling that you talked about in the video, as well as the derivatives, is there anything else that I should consider? Any advice/suggestions would be greatly appreciated, and thank you once again for the video.
Are you looking at military expenditure v time? What is your aim? Keep it focused.
@@MrFlynnIB yes; and my aim was to examine how long it would take for a country such as China to spend just as much if not more than the U.S., as well as the total amount that would have been needed to do so.
@@woreandpiece4815 Yes that could work. Similar idea to when will India's population overtake China's.
@@MrFlynnIB Thank you very much for the feedback; do you also think I should derive the modelling equations and examine those for some extra complexity?
@@woreandpiece4815 I wouldn't recommend it. Depending on the function, it can get really difficult and then really difficult to explain coherently. Let the computer get them but be sure to explain why you chose them. For example what is it about the data points that made you choose a quadratic function?
Mr Flynn, I'm starting my IA shortly and I had a question. My original idea was investigating some tests of convergence for infinite series. I find the topic interesting and I would like to challenge myself into doing a topic outside of the curriculum. My other idea is something to do with modelling crypto (which led me here). What would you say would be my best option
Generally speaking (but not always), IAs that have a real- life application work better. If you can find a good real-life application for your infinite series, this could be very good. But you need a specific aim. 'Investigating different tests for convergence' isn't a good aim as it is too vague. I really like the modelling crypto idea (which is why I made a video on it). Let me know what you decide.
hello sir thank you for the amazing video, i was wondering if it was a feasible idea for me to do my IA on the popularity of a certain book in relation to viral tiktoks made about them. i would use the graph from google trends and find ways to make it more complex. this is for AA HL, it’s all i could think of to get high marks for personal engagement
Not sure that would work although you can collect the data first and see what it looks like. What is your x-axis and what is your y-axis? You are not thinking about personal engagement correctly. You don't have to have a personal interest in the topic to get a high score in the personal engagement criterion. It is more about how you engage with the exploration. For example, I don't have much interest in cricket but I could write an IA with high personal engagement. Conversely, I am very interested in football but I could write an IA with low personal engagement.
@@MrFlynnIB i see what you mean, i was just trying to find a topic that linked to novels in some way.the x axis would be time, the y axis would be the interest. metric on google trends. Alternatively I considered doing the volume of revolution of a cactus plant (a very small one) or a vase but i wasn’t sure what the aim would be. I’m still trying to find a topic that could link to books/reading in some way but i’ve had no luck
@@fade4992 It might work. Get the data and see. The cactus idea would be fine. The aim would just be to get the volume of the cactus. Why you would want to find it is the question.
@@MrFlynnIB lamp shade or cover that disperses the light in a better way or one that would disperse the light equally. Or I could find the surface area or volume of the lamp head thing and compare it to lamps that have been proved to share the most light equally.
@@fade4992 Ya this sounds good.
Hi Mr. Flynn, i am an AI HL student and i was looking for topics for my IA. i am interested in this field and topic but i don't know what all math i will be using in this. Will using calculus to find various points and their significance be enough for an IA?
It depends what you do. Remember the aim is to use mathematics commensurate with the level of the course and to demonstrate understanding. More maths doesn't necessarily mean better. Having said that, you could look at bringing in some statistics to look at what the price might do. Lots of statistics in the AI HL course that you could use.
Hey Mr. Flynn. I am doing AI HL exploration on 'when will China's GDP cross USA's GDP', and my sir told me to use some calculus with modeling. So I wanted to know what calculus can I use for this and how?(my class hasn't reached calculus yet)
Calculus (differential calculus in particular) is all about rates of change. So you can use it to calculate the growth rate of the GDP. For example you could calculate the current growth rate and assume it stays constant to find when they will cross. I like the idea.
@@MrFlynnIB Oh okay, thanks! Also, can I do differential equations too?
@@saumyashah152 It's possible but might get complicated. What I suggested in the previous comment actually involves solving a fairly simple differential equation.
@@MrFlynnIB Thank you!
hey Mr. Flynn I am an AA HL math student and I want to move forward with this Idea however i am struggling with finding math that is in the AA HL syllabus specifically math that is only for math aa hl to implement in this idea could you please help me out
A number of students have asked me similar questions about finding maths that is in the HL syllabus. Remember everything in the SL syllabus is also in the HL syllabus.
how do you model the curve of the span of suspension bridges?
You could use a parabola. Watch my video modelling a skateboard ramp. You could do something similar for a suspension bridge.
hi mr flynn, do you think i can model the amount of covid cases in a specific country as my ia for math ai hl?
Ya sure.
Hi Mr Flynn, I am wondering how maths from the HL syllabus could be applied to this? Because as far as I can see it's largely based on derivatives and SL calc. Thank you!
Remember all that SL calc is also HL calc.
@@MrFlynnIB Thanks for the quick reply Mr Flynn. I know that all SL is within HL, but that is exactly the problem as I believe there is a need to refer to math that is out of SL but within HL? Thanks
@@daniccle You don't have to have maths outside the SL course but you need to demonstrate and better understanding and more sophistication than an SL student would.
This is cool: you could probably go to the wallet address itself and see who moved the Bitcoin
Yes, nice. That's personal engagement right there.
Hey Mr Flynn!! Thanks for all the great video!! I am doing my math AI IA on modelling the force/impact required to make a knockout in martial arts. I'm kind of lost in terms of starting out with the whole thing. I'm panicking and I don't know what to do. It would be great if you could help me sir!!
Have you watched my 8 part IA guide?
Hi Mr. Flynn loved your video. Could i use this topic for math AA HL?
Ya for sure.
Hi Mr Flynn. Can I do something like this with with a country’s GDP growth rate? I’m in Math AA SL.
Ya for sure.
@@MrFlynnIB omg thank you so much for replying. I just wanted to ask if you have any math suggestion I could do. My teacher wants a lot of maths involved, so. Regardless, thank you for helping me out!
@@yoongmoonsahennishuee7970 Lots of maths you could use. Modelling, calculus, solving equations. If your teacher wants more, you can look to add in some statistics.
Hi Mr. Flynn, thank you so much for this video. I was wondering if it'd be a good idea to model the prices of a certain type of cryptocurrency from late 2019 to late 2020 and analyze the impact COVID *possibly* had on the rise and fall of the prices?
Sounds great.
Hi Mr Flynn, I wanted to ask whether the math for this idea is enough for Math AA HL
All depends on what you want to do but sure. You can use calculus.
@@MrFlynnIB could u pls suggest how
@@aamirsangey5976 I talk about this in the video. By calculating the derivative, you can find the max and min points and the points of inflexion. You can also get growth rate at any point.
Hey Mr Flynn, I was just wondering if this would be somewhat suitable for a Math AI HL course, such as modelling the rise in the price of Ethereum. Do you think modelling in this form would be suitable for the HL course?
Ya sounds good. Maybe refine your aim to have a more specific goal. 'Model the price of Ethereum in order to predict its price on 1st of August 2021' is an example.
@@MrFlynnIB Thanks for the feedback Mr Flynn, but what HL topics do you think can be implemented in this form of IA, as isn't modelling SL Based?
@@amanabraham6498 Remember every topic in SL is also in HL. You can just go into more depth for HL.
@@MrFlynnIB Could forms of statistical math also be used in this kind of IA, or would it deviate away from the calculus used in this topic? Thanks for the feedback
@@amanabraham6498 Ya for sure but I would go with one or the other. Keep it nice and concise.
Hi Mr Flyyn, how are you ? I’m having a problem since When I write “Fit” in Geo Gebra no formulas appear to be selected as it occurred in your screen. Do you have any idea why this happens ? Regards !
Strange. Are you in Geogebra classic? Can you share a screenshot?
@@MrFlynnIB After entering again to the web page I was able to do it ! Thank You !
How are you Mr Flynn? When modelling price of bitcoins, there is always the element of randomness when we predict future prices along with information from past pricing data. Thus, how do we incorporate this fact in our IA to make the predictions more accurate? Thank you.
Hi Xuanyi. Of course in reality it is very difficult to predict the price of bitcoin (or anything for that matter). Remember you can also talk about the assumptions you made and limitations of your model. This helps show understanding and reflection. What I'm trying to do in the video is use what has happened in the past to predict the future. You could say something like 'assuming that that the price behaves the same this month as last month, what will the price be?' and then talk about all the problems with that assumption. Let me know if that helps.
@@MrFlynnIB Hi Mr Flynn. This helps a lot. Greatly appreciated. One last question : how can I corporate calculus inside this modelling process? I see you mentioning this in the comment and talked about the point of inflexion. What are the purposes of these? Thanks :)
@@xuanyidu3259 So once you have the function, you can differentiate it to find the rate of growth at different points. You can also find any turning points. The point point of inflection is where the curvature changes. In this example it is where the rate of growth starts to decrease or increase which might be important in your model. I talk about this a bit in the video.
@@MrFlynnIB Alright. Thanks Mr Flynn!! I will try to explore the link between differentiation and this modelling IA!
hi Mr Flynn I love your video. It really helped me make a final decision with my IA. I'm thinking of comparing the prices of ethereum and bitcoin over the years. But I can't seem to think of any math to include in my IA. I'm doing SL AI and so I'm very limited on this. i hope you can help :)
This sounds like a linear regression IA to me. Find the price of Ethereum and Bitcoin at particular times over a given period of time (e.g. 1st of every month for the last 5 years). Create a scatter plot and find the correlation and regression line. Then maybe make some predictions. If Bitcoin goes to 100k, what will Ethereum go to?
@@MrFlynnIB thank you for your feedback!! I'm actually doing exactly that, lol. I've decided to find the correlation between bitcoin and ethereum prices from may 2015 until may 2021, then finding the difference between 2015 and 2021 to make predictions for the next 6 years. I'm not sure if this makes sense yet.
@@Krisy444 Sounds good. Let me know how it works out. I believe the relationship won't be exactly linear. I think Ethereum has actually grown faster in the last 6 years. But that's fine. You can just mention that as you are reflecting.
Hello Mr Flynn I currently study HL AI math. If I do my IA on Modelling the price of Dogecoin in order to predict it's price in one year, how can I apply HL specific topics. Obviously, using calculus to look at the rate of change of price is a way or exponential regression. Can u please help out? How would I structure it out
Both of these ideas seem fine. Let your aim guide you. You could start with exponential regression and see where that takes you. I'd imagine the price of Dogecoin isn't going to grow exponentially for a full year so then you might have to adjust your model. Maybe a logistic model after the exponential one.
@@MrFlynnIB How can calculus be applied to this and is there any context to using calculus in this?, as for HL I aim to use calculus but my teacher said there needs to be context behind it, not just using it for the purpose of the IA. Can you help out with this?
@@danieldjerdjerian362 You can use calculus after you have created the model. Calculus will help you to find the rate of growth of the price of Dogecoin. You can also find the turning points and points of inflexion.
@@MrFlynnIB My teacher said I can do this too but was worried that this was only applicable for the needs of the IA and no bigger purpose, can you explain the context behind using calculus in this case?
@@danieldjerdjerian362 If you want to predict the price in the future, you need to understand how the price behaved in the past and rates of change can help you understand this.
Hi Mr Flynn,
Thank you so much for the video! It was so useful.
I'm thinking about modelling the popularity of rock music but I'm concerned if the topic is maths-y enough? Also, whether that's a strong enough topic for a maths IA for AA SL?
Thanks :)
Sounds interesting. I don't see any reason why it wouldn't be a strong enough topic. Provided the popularity of rock music is changing. First get the data and have a look. If it moves and you can do something like I did with the bitcoin price then great.
Hi Mr Flyyn, thank you for all these great ideas ! I wanted to ask you, how could I develop them into words for my IA introduction ? , could I say something like: “In this investigation I will be modelling the price of Bitcoin, in order to find the function that best fits the model, and then analyse different points where the price of Bitcoin underwent significant alterations (such as a take off or a crash) to understand what events took place in those specific days.
I think you need to make your aim more concise (less words). Maybe 'my aim is to model the price of bitcoin in order to predict its value in 3 months' (or whatever time you want).
@@MrFlynnIB Great, thank you !
@@MrFlynnIB but just one thing, with which of the tools you mention in the video I could develop a possible prediction of the price ? Should I start testing until I find a function that best fits the past prices and seems to continue in a reasonable way ? (Meaning it doesn’t goe drastically down as you showed in the video). And lastly, how could derivatives be applied in the prediction of the price in for example 3 months ?
Sorry for the long questions, and again thank you a lot !
@@tomasbalbal6958 you need to decide this for yourself. Maybe you could make a prediction on the assumption that the next three months will behave exactly like the last 3 months. Then you can find when the next dip in price will occur. Something like that.
@@MrFlynnIB great, thank you a lot !
Your channel is really great. But maybe you can make a better organization of your vids. Put them in playlist in order
Thanks for the feedback. Check out my website, it is very well organized. But yes, you are right maybe I could reorganize the channel.