Explanation of Regression Analysis Results

[jakesmall]

I watched this when I first started studying for my exam and had no clue what you were talking about. Now that I have read all of my lecture notes and come back to this I have found that you explain things in a way that has hit the concepts home more strongly for me. I wish you elaborated a little bit more on all of the aspects but all in all very helpful!
Thanks!

[briannaelms]

How can you explain so clearly in 6 minutes what my lecturers cannot in 2 hours D: Thank you!

[wrjazziel]

Thank you so much, you are so clear, way better than my professors!

[NiniEJ]

Thank you! This was exactly what I was looking for. I had to search through a million videos! You explained it perfectly!

[franciscopintoleite]

Very simple. This is extremely helpful for someone who is starting to study this.

[mattkermode]

You are very welcome. Thank you for the kind words!

[cornaolivier]

Thanks - trying to get my head around stats and this helped so much - perfect, I have looked at a hundred other vids and this was by far the best

[florapannabiro]

This was so much help!!! Thank you =))) Now I got clear the small questions I had in my mind.

[xxxxx]

this is the best thing that happened in my life! thx a lot

[stolenfire]

Made it simple and easy to understand. Especially for Beta calculations using regression!

[mattkermode]

Not by hand, that is for sure. I used a program called StatMac Plus. However, there are numerous regression tools for Excel. SPSS is what I used in university.

These programs simply require you to highlight a time series for your independent and dependent variables. It then produces the analysis in the format shown in the video. All programs return results in a similar format.

Did that answer your question?

[syakeera]

Thank you for the video!
May i ask what does it mean by standard error in each intercept and x? may i know the relation of this standard error to the amazon stock prize and snp500?

[FPC]

Dear Matt,  
Thanks for the video, it helped me a lot. However, I believe you made a small mistake explaining the Standard Errors. They are not the Estimated Y Variance but the Residual's Variance. 
Once again, thanks,  
Best Regards

[mattkermode]

Yes, that occurs occasionally. Attempt to find a rationale for why that might be. It is a good exercise in understanding.

[isabellanocon]

Hey,
thanks a lot for this awesome video! I'm just getting into regression and I'm currently struggling with an exercise I hope you can help me out?
My overall model seems to be significant as I have a R² value of 0, 76 and a F-value way below zero. Coefficient of x1 is 0.44 with a p-value of 0.11 (so not significant) and x2 has a coefficient of 0.23 and a p value of 0, 002 (significant). I am wondering now what this actually means. I guess this results are due to multicollinearity (values of 0, 8) but do I need to exclude now the not significant variable from the model?

[jessinvests]

You’re amazing, thank you kindly! ❤️🙂

[cananseyhan]

PLEASE ANSWER MY QUESTIONS!!!
I have 2 questions:
1)I dont get it that alpha must be zero according to Capm assumption. Because, it accepts efficient market hypothesis. But we can get alpha that different than zero mathematically.
2) In asset pricing model, everybody says we estimate returns. My question is what are we estimate in asset pricing models? What I understand that we use past data of stocks as expected returns to estimate coefficient and after that we look results to say whether the factors used in model significant to explain stock returns. And if the factor significant, what are we gonna get?

[gcinagrahamstheecomedian]

Please help
Question 1 (20 marks)

There is a growing interest among political economists in the factors which influence civil strife (e.g. civil wars, guerrilla wars, terrorism, etc.). Some researchers feel that civil strife may be related to environmental factors (e.g. shortages of clean water and good farmland, deforestation, etc.) as well as political and economic ones. In order to investigate this hypothesis, the data from N = 91 countries were collected on the following variables:

Y = number of lives lost due to civil strife in the last 10 years.
X1 = average annual GDP growth over the last 10 years (measured as a percentage so that a value of, say, 2 means average annual GDP growth of 2%).
X2 = an index of environmental scarcity. (Precise details of the definition of this variable are unimportant; note only that it is a number between 0 and 100 and higher values of this variable indicate more environmental scarcity problems.)
D = a dummy variable which equals 1 if the country is a democracy, and equals 0 otherwise.

Moreover, another variable, Z = X2 ( D, was constructed, and ran a regression of Y on X1, X2, D and Z. Results from this regression are given below in OUTPUT 1.

Use these regression outputs to write a brief report on the effect of environmental, political, and economic factors on civil strife. Your report should include:

Estimated regression equation
A discussion of how coefficient estimates can be interpreted.
A discussion of the statistical significance of your results.
A discussion of what light the results shed on the question of interest (i.e. the effect of environmental, political, and economic factors on civil strife).
Anything else that you would like.

[nathanobregon]

Thanks for the help. I needed this....

[jimbosloan]

Hi there! thanks a lot for the video, it's really easy to understand. I'm doing a somewhat related research right now on Hang Seng index. Would you mind suggesting any sources that you have taken the data from, please.