Simple Linear Regression - ANOVA
Вставка
- Опубліковано 23 лип 2024
- In this video I explain how partitioning the variability and the normality assumption yield an F test for a simple linear regression. Additionally, I provide an example in Excel. Here is the link to the data used in this video example: drive.google.com/file/d/1dU_4...
Great video! Specially for people who need a refresher on the concepts years out of school :)
You just saved my semester... Thanks a lot
Very helpful video! Nice explanation!
clear explanation, THANK YOU
I like yours way to explain it
❣❣❣
Thank you SO MUCH
Hi Michelle,
Thanks for such a great explanation!....can you pls explain the degrees of freedom part again or make a video on it!
Thank you for you efforts!
Glad you found the video helpful! In general, the degrees of freedom are the number of values that can freely vary. For example, the degrees of freedom "Total" is n - 1. This is because the degrees of freedom "Total" is for the calculation of the sum of (yi - ybar)^2. If I know ybar, and I know y1, y2, ... yn-1, I do not need to know yn because I can find yn using the sample mean, ybar. Therefore, for the calculation of the sum of (yi - ybar)^2, there are n-1 freely varying values. If you have a follow-up question to this, or if I misunderstood the question, please let me know. There is a lot to discuss when it comes to degrees of freedom :-)
@@Stats4Everyone Great explanantion!!Thank you so much for your response and such an amazing content!
@@Stats4Everyone hi, I have misunderstood the term degrees of freedom too. Can you explain why we consider that we have already known y(bar) when calculating degree of freedom? I think y(bar) is the average of y in n-observations, so y(bar) should be random like y, then we cannot known y(bar). Btw thanks for your video.
@@namhainguyen6541 The degree of freedom "Total" , is the number of variables that can take on any value in this equation (yi - ybar)^2 , in this equation we need to estimate ybar first , so ybar = sum(yi)/n , after calculating this we lost the ability to freely change one value of all values of y, that is because one value has to adapt for the change in other values , for example if we have a variable that is like this {2,4,1,5} when calculating the mean it is equal 12/4 = 3 , so after we calculated ybar here, the question is how many values in the set can be changed while having the same value of ybar ? , only 3 of them can change and the last one must adapt to that change to have the same ybar. or in other words to have sum(yi-ybar) add to zero .
so if we didn't estimate the mean and we had the population mean , we can change all 4 values freely and we won't loose any degrees of freedom because all the values in the set don't need to add up to zero .
@Stats4Everyone Hi Mam, Great explanation
Great video, however I got 6 or 7 ads during the video ... Unbearable.