David, can you go a bit more into detail on what your results are telling you? For example, I am a bit confused, as you don't indicate the market cap of GOOG, so how do you know how the SMB factor is affecting it?
Great content, however, once you run the regression you get an R Squre of 0.34 which indicates that the analysis don't capture the full risk of the portfolio (66% not explained).
Dear Richey Sir, This is a very helpful video. Thank you for this. Sir, you have calculated returns by using : (Pt/Pt-1)-1. Can we use log returns i.e LN(Pt/Pt-1). Which method is best for daily, monthly and annual returns? Please help me.... Thank you Priya
Amazing video, but you should have spent some time interpreting the p-values of the factors. Not all the factors are statistically significant for every ticker symbol. The good news is that in 2022 we now have portfolio optimizer haha
how do we find out the betas and the error term In the fama-french regression? because i have a problem to find the systematic risk and idiosyncratic risk in the regression. after that how we supposed to do after we find the betas and the error term to have the idiosyncratic return? thank you
@David Richey Thank you for the video. How could i have the excel access about the computations of the factors (size, BM and momentum). I am working to build a carhart model for african market in which i need to construct the mimicking portfolio since in the webpage thy never have done it. Any help please?
It is a good video, thanks. If u can tell more about the meaning of function like (linest(y,x,1),1,3) can be better, because I don't know why they show put on like this.
Hi David, Thanks for this video. I've a portfolio of multi-currency stocks. So I have stocks in CHF, EURO, and GBP, along with USD. The first stock, I tried the regression with is a CHF stock. I got p-values around 0.18 and higher. Which makes me to believe that this isn't a good model to use for this stock. Is my interpretation correct?
well, first of all i would like to say it is very helpful video. However, i have a question that suppose if i want to calculate on rolling window basis Fama French 4 factor Model for Malaysian stock (Mintye Industries Bhd (5886.KL). Still i will use/ consider Mkt-Rf, SMB,HML,MOM?. My second question is related to Beta(s).... in this which one is Market beta and which one is industry beta?
Fama-French isn't particularly useful over the short-term, other than for performance attribution (e.g., "to which factor can we attribute the portfolios out/underperformance to?"). Weekly is likely the shortest time frame you'd want to use; higher frequency data (daily) is usually noisy and too sensitive to asynchronism across factors.
Hi. Thank you for the usefull video. I have a question: Is it necessary to take the Risk Free Rate included in the Farma French Data or is it also possible to use a different Risk Free Rate where it is still possible to use the other Colums of the Farma French Data. I would appreciate an answer!
thanks for sharing. but i have some questions .could you please help to explain? 1.i still can't understand why we need to /12 to get monthly data in the end for three factor loading are monthly data. 2.when using the linest function, how could we get the T-TEST or p-value of the coefficient ? Now i am can calculate do the fama-french model for a bundle of funds, i did get the coefficient of each factor loading, but i have no idea whether they are significant. I'm appreciated your attention.
Is it necessary to take the excess return? In my classes we always used for the first beta the regression from the return to the market return without risk free
Technically it is. If you look at the Fama French definition, the first variable is return of the market minus the risk free rate, ie the excess return. The SMB and HML measure historic excess returns of small caps over big caps, once again using the excess return. That being said, currently the risk free rate is so close to 0, it would essentially give you the same answer. Looking at column (letter) O on the Excel, almost all of the values are 0.
Hey thanks for the video!! I just wanna ask you if it's possible to do the same calculations with Japanese equities, like is it possible to use Nikkei 225 or TOPIX instead? And if so what alterations do I need to do regarding the Farmer French Factors? Can I apply the FF3 data you used in the video for Japanese equities or do I need to get it elsewhere? I really need your help!
Hi, You can use any kind of data and you can apply fama french model over that. You can also use fama french daily data / monthly data / weekly or yearly data.
I think that's the best Fama-French video tutorial on yt. Well explained, thanks!
You sir, are a life saver.
David, can you go a bit more into detail on what your results are telling you? For example, I am a bit confused, as you don't indicate the market cap of GOOG, so how do you know how the SMB factor is affecting it?
you have to multiply the average return of each portfolio by its proper beta to get the cost of equity
Thank you so much for your help. Would you please also make the same calculation on Rstudio?
YOU SAVED MY LIFE! THANK YOU!
Great content, however, once you run the regression you get an R Squre of 0.34 which indicates that the analysis don't capture the full risk of the portfolio (66% not explained).
Hey, Sir David. Thanks for you this video. and Have you a video about how to sorted the stocks into some portfolios by size (Fama French) sir?
Dear Richey Sir,
This is a very helpful video. Thank you for this. Sir, you have calculated returns by using : (Pt/Pt-1)-1. Can we use log returns i.e LN(Pt/Pt-1). Which method is best for daily, monthly and annual returns? Please help me....
Thank you
Priya
Yes
Amazing video, but you should have spent some time interpreting the p-values of the factors. Not all the factors are statistically significant for every ticker symbol. The good news is that in 2022 we now have portfolio optimizer haha
Is it true that market returns are negative in some companies?
Can we use the same method for calculating it for cryptocurrency as well?
Why do you use simple returns and not log returns?
You just made my grade go up!!! Thank you so much!! 💋
this video really helped me thank you very much
The video is very helpful, do you have a blog? Thank you
Hi, Thanks for the useful lecture. Why did you multiple the factors by their averages? Did you consider averages as betas?
Thank you so much! was a life saver
You're a legend bro, thanks.
randomloverofcheese thanks lol
Liked the video. Fairly nice explanation. Thanks
how do we find out the betas and the error term In the fama-french regression? because i have a problem to find the systematic risk and idiosyncratic risk in the regression. after that how we supposed to do after we find the betas and the error term to have the idiosyncratic return? thank you
@David Richey Thank you for the video. How could i have the excel access about the computations of the factors (size, BM and momentum). I am working to build a carhart model for african market in which i need to construct the mimicking portfolio since in the webpage thy never have done it. Any help please?
It is a good video, thanks. If u can tell more about the meaning of function like (linest(y,x,1),1,3) can be better, because I don't know why they show put on like this.
You are life saver!!!! Thank you!!!!! :D
Shahzeb Muhammad glad to help!
Would i be able to use Carhart’s Four Factor Model data instead of the fama-french three factor model?
How do we interpret the results? Is that the expected return for the next year?
no its the risk-adjusted return from the given data, so that is the minimum return an investor would have needed.
Paul Glasmacher needed? To make the investment worth the risk? Thanks!!
@@silmm1886 you are welcome
guys, im just wondering, where i can get instant financial ratios without having to go to their balance sheet, u guys have any info? thx so much
So when it's to calculate FF's five-factor model, the only change is the number from 1,1,3 to 1,1,5? is it correct? Thank you!
I have the same question , have you find if your statement is correct ?
haha you need to have data for the other two factors
Hi. Yes I checked this. This works for fama french 5 factor model as well. U need 5 factors data,
what's the last step for? my teacher says the abnormal return will be measured by jensen's alpha and t-stastistic will be given in the regression.
Thank you really useful bro
Hi David,
Thanks for this video. I've a portfolio of multi-currency stocks. So I have stocks in CHF, EURO, and GBP, along with USD. The first stock, I tried the regression with is a CHF stock. I got p-values around 0.18 and higher. Which makes me to believe that this isn't a good model to use for this stock. Is my interpretation correct?
well, first of all i would like to say it is very helpful video. However, i have a question that suppose if i want to calculate on rolling window basis Fama French 4 factor Model for Malaysian stock (Mintye Industries Bhd (5886.KL). Still i will use/ consider Mkt-Rf, SMB,HML,MOM?. My second question is related to Beta(s).... in this which one is Market beta and which one is industry beta?
Why do you use the arithmetic average from minute 12 onwards and not the geometric mean?
Many thanks!!! Really appreciate your video!
can you please make a video on 25 portfolios double sorted and single sorted
Hi Umar. U find a way to do this?
is that the same way for short-term analysis? just use daily data instead of?
Fama-French isn't particularly useful over the short-term, other than for performance attribution (e.g., "to which factor can we attribute the portfolios out/underperformance to?"). Weekly is likely the shortest time frame you'd want to use; higher frequency data (daily) is usually noisy and too sensitive to asynchronism across factors.
is it really monthly return at 6:45
Hi. Thank you for the usefull video. I have a question: Is it necessary to take the Risk Free Rate included in the Farma French Data or is it also possible to use a different Risk Free Rate where it is still possible to use the other Colums of the Farma French Data. I would appreciate an answer!
i also have the same que. is it possible?
thanks for sharing.
but i have some questions .could you please help to explain?
1.i still can't understand why we need to /12 to get monthly data in the end for three factor loading are monthly data.
2.when using the linest function, how could we get the T-TEST or p-value of the coefficient ? Now i am can calculate do the fama-french model for a bundle of funds, i did get the coefficient of each factor loading, but i have no idea whether they are significant.
I'm appreciated your attention.
I'm not aware of a formula to dynamically show P-value; there is an R square one though (=RSQ).
Is it necessary to take the excess return? In my classes we always used for the first beta the regression from the return to the market return without risk free
Oh and btw, great video of course!
Thanks!
Technically it is. If you look at the Fama French definition, the first variable is return of the market minus the risk free rate, ie the excess return. The SMB and HML measure historic excess returns of small caps over big caps, once again using the excess return. That being said, currently the risk free rate is so close to 0, it would essentially give you the same answer. Looking at column (letter) O on the Excel, almost all of the values are 0.
Thank you so much !! you saved my life
Where is the value of beta? is it not required to calculate the beta? i don't understand when comparing it with the actual formula of FamaFrench
Hey thanks for the video!! I just wanna ask you if it's possible to do the same calculations with Japanese equities, like is it possible to use Nikkei 225 or TOPIX instead? And if so what alterations do I need to do regarding the Farmer French Factors? Can I apply the FF3 data you used in the video for Japanese equities or do I need to get it elsewhere? I really need your help!
Hi,
You can use any kind of data and you can apply fama french model over that. You can also use fama french daily data / monthly data / weekly or yearly data.
Very nice!! Keep it up. Thanks!!
Will do, thanks!
Hi, could you explain what is meaning of your results? I mean what is meaning of 10.73% cost of equity?
This stock will give you a yearly return of 10.73% to compensate the risk they take investing their capital (i think lol)
how do u calculate SMB, HML urself?
Thank you my dude!!
Great video, thank you so much!
THANK YOU SO MUCH
Legend thank you
That was amazing. Thank you so much!!
Awesome video.
Great video!
Thanks!
You not calculate the essential SMB and HML portfolios ....
Thank you for explaining FF3 in Excel!
You are welcome!
are the x variables the beta ?
yes
Thank you a lot!
Good job
thank you champion
How can we calculate alpha, please?
OMAR Mamlouk alpha = realized return - expected return from CAPM
OMAR Mamlouk alpha = realized return - expected return from CAPM
so nice!so nice!so nice!so nice!so nice!so nice!so nice!so nice!so nice!so nice!so nice!so nice!so nice!
amazing
THanks
Genial !