-
Notifications
You must be signed in to change notification settings - Fork 3
/
My plan
38 lines (24 loc) · 1.53 KB
/
My plan
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
My plan
General questions:
What is this person trying to do, and why?
I am trying to evalute the impact of economics growth on the higher education expansion. For example, will higher education is likely to occur in a county with higher foreign direct investment per capita?
What have they done so far?
So far I have otained the dataset (both higher education dataset and economics dataset) and merged by zipcode for China's higher education expansion study;
What are they working on?
I am not sure if my merge makes sense(I merged twice, the first time merged by Zipcode, the second time to merge by population1990). It looks a little bit messy.Needs to take a deep into it.I am not sure how to do the merge (many-many)in R. I did the merge many to many using Stata. But I prefer to do it in R.
How much progress are they making? Have they hit a snag?
Have they left out something important from consideration?
The next step:
This week:Check the merge results;
Next week: primary analysis;
<<<<<<< HEAD
=======
Update on March 19.2017
I am checking the potential clustering solution.
Kmeans did not work well, But the complete/single/ ward methods works well
>>>>>>> a32cb3731b5cdd09bb11a56e3da690e2e471b025
update on April 17,2017
I have fixed checked the merge results. It is very good. Cannot find nothing wrong;
Prefecture analysis---Both OLS with fixed effect and Logit model fits well.
Urban analysis- Same thing. These results/findings are quiet different.
Questions/Potential concerns for this week: Fit both model with the weight variable.