#StackBounty: #r #logistic #survival #recurrent-events (Repeated) Event Model – How to model entry?

Bounty: 50

I am modeling the entry of 30 firms into 25 industries over time. The unit of analysis are industries.

I have cross-sectional panel data on the industries and know when a firm enters. I have 5 industry-level co-variates that change over time.

I would like to know which of these 5 co-variates best predicts the entry of a firm into the industry.

Note: In some industries multiple firms enter (at different points of time).

What type of model do I use?

• A logistic panel model with fixed effects for industry (and year)?
• An extended survival model? Perhaps a Cox regression that models repeated events?
• Or something different altogether?

I am looking for a statistical approach to be performed in R.

Example data:

``````dt <- data.table(firm = c(NA, "F1", NA, "F2",NA, NA, NA, "F1", NA, NA, NA, "F3","F4", NA, NA, NA),
industry = c(1, 1, 1, 1, 2, 2, 2, 2, 3, 3, 3, 3, 4, 4, 4, 4),
year = c(2000, 2001, 2002, 2003,2000, 2001, 2002, 2003,2000, 2001, 2002, 2003,2000, 2001, 2002, 2003),
entry= c(0, 1, 0, 1, 0, 0, 0, 1, 0, 0, 0, 1, 1, 0, 0, 0),
x1 = c(runif(16)),
x2 = c(runif(16)),
x3 = c(runif(16)),
x4 = c(runif(16)),
x5 = c(runif(16)))

> dt
firm industry year entry        x1         x2        x3         x4         x5
1: <NA>        1 2000     0 0.8709432 0.53087423 0.8454785 0.25285903 0.05671375
2:   F1        1 2001     1 0.2775248 0.57784857 0.3895359 0.29725940 0.17875968
3: <NA>        1 2002     0 0.4863695 0.44049405 0.6658640 0.46859840 0.72456472
4:   F2        1 2003     1 0.4604883 0.47147933 0.1241863 0.30133078 0.88424245
5: <NA>        2 2000     0 0.8946535 0.41725531 0.7416106 0.47860660 0.15310571
6: <NA>        2 2001     0 0.4675261 0.17393733 0.2680798 0.24721979 0.25179696
7: <NA>        2 2002     0 0.1801262 0.76724741 0.3157204 0.20496050 0.21098053
8:   F1        2 2003     1 0.9967714 0.54301656 0.9133624 0.41144085 0.39097097
9: <NA>        3 2000     0 0.4584210 0.26767502 0.4698928 0.62344696 0.96677505
10: <NA>        3 2001     0 0.5241493 0.21622151 0.7802806 0.89008628 0.28688089
11: <NA>        3 2002     0 0.1472759 0.58341162 0.5209162 0.42375726 0.06895383
12:   F3        3 2003     1 0.4255832 0.09506578 0.4067654 0.25417083 0.32360013
13:   F4        4 2000     1 0.1922379 0.61931944 0.3609198 0.99608796 0.57967160
14: <NA>        4 2001     0 0.2681883 0.26763619 0.6279573 0.57704165 0.07834483
15: <NA>        4 2002     0 0.2431978 0.69426770 0.4675626 0.77913438 0.87128792
16: <NA>        4 2003     0 0.7301488 0.94025212 0.5586453 0.05350145 0.98182507
``````

Get this bounty!!!

This site uses Akismet to reduce spam. Learn how your comment data is processed.