I think I got a better grasp on understanding the outputs. I have a LOM due in less than 12 hours and I am just beginning to understand the basics in regression. So thank you for explaining.
thank you so much for this lifesaver i am currently studying in year 11 with no prior knowledge of regression, and my crazy employer thought it would be a good idea to make me learn all this on my own
Thanks for the video! Is there an intuitive explanation why the degrees of freedom of the regression (SSR) is just one whereas the df of SSE is n-2? In the end, we use the same number of observations (16)
Hey! Great video! I know this is from a long time ago.. But I'm hoping you can help me. If I have a multiple regression table, and on their own, each independent variable is HIGHLY significant to the dependent variable according to the p-value and R2. But when I put them all in one table, the p-value of 1-2 predictors fall outside of the 0.05 limit to an extreme 0.35. However, my R2 increases to .995. Should I eliminate those variable even though on their own, they're significant? Or how should I interpret these results?
Good to see you showed the equation. I don't understand anything. Just to get the point Mr. Like what values show that hypothesis wrong or right and why.
outstanding example; thank you; it's hard to find a clear example where multivariable regression passes all statistical significance tests and explains variation better. I wonder if the best additional (3rd) variable might be fielding percentage??
Hi jason.. can you please do a video on ms excel solver.. If i have a multiple regression model.. with 6 independent variables (X1 to X6) and known value of Y.. so we fix the value of Y and find out combinations of those 6 parametera that when equation solved,give close result to our value of Y. I am doing it for calibration of my simulation data.
I assume you mean from a logistic regression with a binary dependent variable. To my knowledge, there is no canned way to do this, but you can build a log likelihood function and use Solver to iterate to a solution. Other programs, such as R or Stata can do this more easily.
R^2 is the proportion of variance accounted for by the regressors divided by the total variance in y. As you add regressors (MLR), the numerator increases. I hope that helps!
Excel does that "automatically" when you use the analysis toolpak, but if you need to calculate it you can use =LINEST on the original data set. You need the original data, and then select an array of cells to put the output, use =LINEST(yvars,xvars), and then press CTRL-SHIFT-ENTER (it's an array formula).
Mr. Jason, pls speak up louder and with more firmness in the words, it's difficult to pick up by gawking at thr screen as to wht you're saying No offense 🙂
Sorry Jason, I am not happy with your interpretation of the regression output. I was expecting clearer interpretation. Professionally speaking, you should start by explaining the objective of regression for explaining the relation between variables. Then, you start by explaining each out put and what it means in a simple way. What happened is you explained vaguely with mentioning "this kind important" etc. I was hoping for more bro.
I mean, fair enough. Nonetheless, if you want to understand the objective of regression, I would recommend watching the many other videos I have on regression. I just recorded a new video on output today, and I am confident it fails to address your concerns, because that's what my "Regression - Overview" videos do. To each their own, I guess. Check out my dummy interpretation videos if you really want to get into the weeds.
I've literally understood almost nothing from this video. If you don't want to be open to explain things in details for others to learn, please no one's forcing you to do a video. 🤦