Get my FREE cheat sheets for Public Health, Epidemiology, Research Methods and Statistics (including transcripts of these lessons) here: www.learnmore365.com/courses/public-health-epidemiology-research-methods-and-statistics-resource-library
I'm watching halfway. I just hit subscribe. The content you put here in this video is just so well-explained! You translate codes into layman's term and have "tidily" edited your video! I love the zoom in and out effect of it and the sound effect. Not too much. Just right. Not annoying, rather impressive. Thank you for sharing your knowledge to us, Greg!
Greg, thanks for ALL your elaborate videos and the structure of the lessons. In addition, the way you explain the code methodically! Love it. I was so stressed about replacing NA with none for the variable, gender (Pt 3 of handling missing values), turns out the variable is sex. Phew
Please do a video on imputation in R! I was working on something and I was confused as to whether my data was "missing at random" or another option so I wasn't sure how to handle imputation.
Could you please make a video on testing MCAR and, given its assumption of multivariate normality, talk specifically about what to do with factor variables or logicals?
Hello Greg, I have a question. I ran the following code and i want to run a regression on the adjusted dataset now. However, it takes the unadjested dataset instead of the adjusted one. Also, it creates a new dataset called ''.'' (so just a dot). This dataset is the correct adjusted one, but I cannot even use it. I am confused. library(dplyr) library(ggplot2) library(tidyverse) iabbd_8010_v1%>% select(Destination, Year, Origin, Mstock_Total, Mstock_Low, Mstock_Med, Mstock_High, Distance, Democracy_origin, Democracy_destination, GDPpc_origin, GDPpc_destination, Language, Population_origin, Population_destination, Border)%>% mutate(Mstock_Total = replace(Mstock_Total, Mstock_Total == 0, NA))%>% drop_na(Mstock_Total)%>% mutate(Mstock_Total = log(Mstock_Total))%>% mutate(Population_origin = log(Population_origin))%>% mutate(Population_destination = log(Population_destination))%>% mutate(Distance = log(Distance))%>% mutate(GDPpc_destination = log(GDPpc_destination))%>% mutate(GDPpc_origin = log(GDPpc_origin))%>% View() reg1 = lm(Mstock_Total ~ Distance + Language + Border, data = iabbd_8010_v1) --> so actually i should do data = . but it says ''.'' doesn't exist summary(reg1)
Hello, thank you for these videos. They are very helpful. Is there a video on what program evaluation is and how that looks in the global health context?
🖐 great video, thanks. But didn't work for my case. There is a char format column, in my table (14 columns * 50000 rows) with up to 7000 missing values, but na.omit() can't find them. Is it possible it's due to invisible typed "space" that na.omit() can't find them? I hope I was clear.
Greg, I am having trouble seeing the difference between changing missing data to value vs imputation. Are they not the same? Can you explain the difference. Thanks! Great lessions by the way.
Drop_na, complete.cases worked perfectly on R studio . But when I write the same code in kaggle new data frame doesn't have any value ?? Any suggestions ??
Great vid but instead of using the "%>%" function, how could we have done it? Since we are not able to save these changes made to the original dataset using "%>%" function.
Hi Greg love your videos! Im a medical student who is going to intercalate next year in public health which im very excited about. Ive got a choice however between MSc International public health (with a focused stream on humanitarian studies) or MSc Humanitarian studies. Im interested in the working humanitarian relief space, but im wondering if I should I keep my studies a bit broader at the moment and study the MPH. Would be interested to know what you think in terms of if one would be more advantageous in my career. thanks James
Great introductory video! Thanks! :D I have a question for everyone: I'm imputing missing values for Gender in a dataframe. Out of the complete rows (no NAs) Male=61.89% and Female=the rest obviously. Is there a way I can impute the values randomly but in these proportions? It feels like there must be but I am new to R... Thanks!!
This video has useful information. However, it didn't help me understand missing data. It helped me understand how to filter out or replace missing values with a constant. Not the same.
Get my FREE cheat sheets for Public Health, Epidemiology, Research Methods and Statistics (including transcripts of these lessons) here: www.learnmore365.com/courses/public-health-epidemiology-research-methods-and-statistics-resource-library
You have saved hundreds if not thousands of hours of beginning analysts time. Thanks!
You're welcome!
Hello sir, this is amazing. You're a wonderful teacher. Please do more. Very many thanks from me here in Kenya
Thank you very much for the feedback. I’ve been to Kenya. Lovely country.
I have been having problems with functions, can you help? I would appreciate so much
I'm watching halfway. I just hit subscribe. The content you put here in this video is just so well-explained! You translate codes into layman's term and have "tidily" edited your video! I love the zoom in and out effect of it and the sound effect. Not too much. Just right. Not annoying, rather impressive. Thank you for sharing your knowledge to us, Greg!
I just finished watching and taking down notes. Huge applause to you, Greg!!!
What awesome feedback, thank you! I really appreciate it!
I know this video is old, but still very helpful! I love your channel, you make stats and R fun :D Thanks for making these, keep up the great work.
Glad you like them!
Thank you so much Greg, can you please tell me what software you use for video editing?
Thanks in advance
You said, "Boom Shakalaka" LOL! Most awesome video ever.
Great video. Looking forward to your videos about imputation and the MICE package. Keep’em coming!
Best r tutorial , visuals, pace, delivery....so good!
Wow, thanks!
Greg, thanks for ALL your elaborate videos and the structure of the lessons. In addition, the way you explain the code methodically! Love it. I was so stressed about replacing NA with none for the variable, gender (Pt 3 of handling missing values), turns out the variable is sex. Phew
This is a very insightful explanation:) thank you!
Glad you find it insightful. Thank you
great way of yours to finally simplify stats ...thank you
Glad it was helpful!
Please do a video on imputation in R! I was working on something and I was confused as to whether my data was "missing at random" or another option so I wasn't sure how to handle imputation.
Thanks for the help, really appreciate, I have exam tomorrow, and you really helped Sir.😃❤
Glad it was helpful! Thank you :)
11:48 - " Take care, stay well, don't do drugs, always do best, speak to you soon. Bye! " - that's a cool outro
Impressed!
Thank you. Cheers
Great video, helped me a lot cleaning some datasets in an easy way.
supper video
clear,
thank you soo much
You are a good teacher i like your video
Waoo...another great video
Could you please make a video on testing MCAR and, given its assumption of multivariate normality, talk specifically about what to do with factor variables or logicals?
Hello Greg,
I have a question. I ran the following code and i want to run a regression on the adjusted dataset now. However, it takes the unadjested dataset instead of the adjusted one. Also, it creates a new dataset called ''.'' (so just a dot). This dataset is the correct adjusted one, but I cannot even use it. I am confused.
library(dplyr)
library(ggplot2)
library(tidyverse)
iabbd_8010_v1%>%
select(Destination, Year, Origin, Mstock_Total, Mstock_Low, Mstock_Med, Mstock_High, Distance, Democracy_origin, Democracy_destination, GDPpc_origin, GDPpc_destination, Language, Population_origin, Population_destination, Border)%>%
mutate(Mstock_Total = replace(Mstock_Total, Mstock_Total == 0, NA))%>%
drop_na(Mstock_Total)%>%
mutate(Mstock_Total = log(Mstock_Total))%>%
mutate(Population_origin = log(Population_origin))%>%
mutate(Population_destination = log(Population_destination))%>%
mutate(Distance = log(Distance))%>%
mutate(GDPpc_destination = log(GDPpc_destination))%>%
mutate(GDPpc_origin = log(GDPpc_origin))%>%
View()
reg1 = lm(Mstock_Total ~ Distance + Language + Border, data = iabbd_8010_v1) --> so actually i should do data = . but it says ''.'' doesn't exist
summary(reg1)
Hello, thank you for these videos. They are very helpful. Is there a video on what program evaluation is and how that looks in the global health context?
🖐 great video, thanks. But didn't work for my case.
There is a char format column, in my table (14 columns * 50000 rows) with up to 7000 missing values, but na.omit() can't find them.
Is it possible it's due to invisible typed "space" that na.omit() can't find them?
I hope I was clear.
I like the audio quality
Thank you!
Greg,
I am having trouble seeing the difference between changing missing data to value vs imputation. Are they not the same? Can you explain the difference.
Thanks!
Great lessions by the way.
Drop_na, complete.cases worked perfectly on R studio .
But when I write the same code in kaggle new data frame doesn't have any value ??
Any suggestions ??
Dear Greg, I've been watching all you R video in your other channel " R Programming 101". Why didn't you put this R video in that channel?
can we replace the NA without using library?
Great vid but instead of using the "%>%" function, how could we have done it? Since we are not able to save these changes made to the original dataset using "%>%" function.
create a new dataframe
Why do you use pipe at the end of each command?
I want videos on text manipulation
Hi Greg love your videos! Im a medical student who is going to intercalate next year in public health which im very excited about. Ive got a choice however between MSc International public health (with a focused stream on humanitarian studies) or MSc Humanitarian studies. Im interested in the working humanitarian relief space, but im wondering if I should I keep my studies a bit broader at the moment and study the MPH. Would be interested to know what you think in terms of if one would be more advantageous in my career. thanks James
if you know please upload a video for the matlab code of Multivariate imputation by Chained Equation(MICE)
Great introductory video! Thanks! :D
I have a question for everyone: I'm imputing missing values for Gender in a dataframe. Out of the complete rows (no NAs) Male=61.89% and Female=the rest obviously. Is there a way I can impute the values randomly but in these proportions? It feels like there must be but I am new to R... Thanks!!
im a bit late but i guess if you do a conditional on a random number generator. So 0-0.3811 is Female and 0.3812+ is male
Na_if( ), it is just what I am looking for.
Thank you for the feedback, Reddy. Hope all is well
Why my latest R version shows that no tidyverse package 😫
This video has useful information. However, it didn't help me understand missing data. It helped me understand how to filter out or replace missing values with a constant. Not the same.
Sound system is very poor
wa la
*what! "Don't do drugs"?? , youtube is one of the most addictive drugs in the world.