Hi Guru, I was hoping you could help me out with some regression testing and correlations. Attached is a word doc with the requirement steps and some graphs that I've already compiled. There is also a separate dataset used for the graphs. Let me know if you have questions...
hi would you like me to help you? is it simple or multiple regression? also where is the word file where data can be found? I think you forgot to include the data file. please reply asap. Thanks
This is a simple regression analysis using a single explanatory variable. I'm pretty sure the point is to analyze all and indentify which is the best to use.
Attachment: 2012-06-23_180804_regression.doc
Attachment: 2012-06-23_180834_csdata.xls
Welcome and Thanks for requesting me.
That's too big an assignment, asking for so many things. Do you need only specific parts of that? Let me know.
Thanks.
Too big for the price, or just in general?
Thanks,
Too big for the price. Moreover, in the data file, what do the variables HSM, HSS, HSE, SATM and SATV mean?
You know, there isn't any sort of explanation of the variables. My thinking is that it's High School Math, Science and English scores and SAT scores Math and Verbal. Based on the scatterplots, it seems that the Math and Science scores would be the best predictors.
How much of a bonus would you require to help with HSM and HSS? I'm really stuck on the steps below.
Generate correlations and covariances between variables of interest.
Examine results for potential linear relationships and identify models: response = constant + slope * predictor.
Perform regression analysis.
Examine results of regression analysis for explanatory power and statistical significance.
Examine error analysis results for compliance with assumptions.
So, the independent variables to be used are HSS and HSM only?
That's what I'm going with. They appear to have the most obvious positive trend.
Ok, I will do a multiple regression with GPA as y and HSS and HSM as x1 and x2. I request for a $20 bonus. By when is this due?
Thank you! Hang on a couple minutes and I'll post what I've compiled so far. I mostly need your help to interpret it, unless of course you see something terribly wrong. Then please point out!
Ok, I am here.
Alright. Here's what I'm working with. I don't think anything's wrong, per say. I just need some assistance with explaning the calculations.
Attachment: 2012-06-24_143905_regression.doc
HiYou seem to have all the work already. I went through your work. Its good. Since the p- value is 0 for HSM, HSM is a significant predictor in the multiple regression model.But p- value for HSS is greater than 0.10, which means HSS is not a significant factor in the regression.But R^2 value of 20% indicates that the model is poor and forecasts made using this model may not be reliable.I would suggest keeping all the variables in the model (leaving out sex) and then running the regression to see if we get a stronger model. I can't say what we might get but may be we should.If you think we should, let me know. I can help with that.I hope that helps. Let me know if you need any clarifications on my answers. I will be happy to provide them. Else, please give a rating for my answer. Thanks.
Experience: I have a Master's degree in Engineering and a very wide teaching experience of more than 25 years at various levels.
Hi
I just ran your model with all other variables included. Its the same. Here is the output ...
Only HSM appears to affect the GPA strongly. So you can keep what you have done.
I hope that helps. Let me know if you need any clarifications on my answers. I will be happy to provide them. Else, please give a rating for my answer.
I got the same thing. I guess I'm just overthinking this. Thanks for looking this over. Do you stlil require the $20 bonus?
No, you can just accept the answer. Bonus is not required since I didn't do much.Thanks.