Hi all,
I am looking to interpret the changes of my panel data fixed effects model using quantile regression at the 25th, 50th and 75th percentile. First off, is -xtqreg- the right command for this?
The original model is (1), models 2, 3 and 4 are quantile regression at the 25th, 50th and 75th quantile respectively.
Could somebody please assist with interpretation. To me it looks like this table has little added value to my report as from SIZE to WOB are all control variables. Only ERS and EGPI become insignificant. What could be an explanation for that?
My supervisor is asking for robustness check using quantile regression, however I have never worked with this before.
Best,
Peter
I am looking to interpret the changes of my panel data fixed effects model using quantile regression at the 25th, 50th and 75th percentile. First off, is -xtqreg- the right command for this?
The original model is (1), models 2, 3 and 4 are quantile regression at the 25th, 50th and 75th quantile respectively.
Could somebody please assist with interpretation. To me it looks like this table has little added value to my report as from SIZE to WOB are all control variables. Only ERS and EGPI become insignificant. What could be an explanation for that?
| (1) | (2) 0.25 | (3) 0.5 | (4) 0.75 | |
| w_ROE | w_ROE | w_ROE | w_ROE | |
| EGPI | -.637** | -.753 | -.625 | -.522 |
| (.297) | (1.865) | (1.054) | (1.068) | |
| GW | .058 | .224 | .042 | -.104 |
| (.435) | (2.63) | (1.487) | (1.506) | |
| ERS | .027** | .039 | .026 | .015 |
| (.012) | (.078) | (.044) | (.045) | |
| SIZE | -2.482*** | -1.091 | -2.619 | -3.852 |
| (.74) | (4.199) | (2.374) | (2.404) | |
| w_LEV | -15.603*** | -28.378** | -14.343* | -3.017 |
| (2.603) | (14.268) | (8.075) | (8.17) | |
| w_GROW | .188*** | .209** | .186*** | .167*** |
| (.013) | (.095) | (.054) | (.055) | |
| w_CAP_INT | 49.392*** | 47.734 | 49.555** | 51.025** |
| (7.2) | (43.708) | (24.705) | (25.018) | |
| w_RD_INT | -63.428*** | -62.962 | -63.474 | -63.888 |
| (23.418) | (114.638) | (64.797) | (65.618) | |
| BSIZE | -.012 | .05 | -.018 | -.073 |
| (.066) | (.414) | (.234) | (.237) | |
| BINDP | 1.979** | 1.743 | 2.002 | 2.211 |
| (.879) | (5.229) | (2.955) | (2.993) | |
| CSRCOM | .239 | .012 | .261 | .463 |
| (.464) | (2.856) | (1.614) | (1.635) | |
| CEOD | -.784* | -.821 | -.78 | -.748 |
| (.417) | (2.585) | (1.461) | (1.48) | |
| WOB | -1.594 | -2.755 | -1.48 | -.452 |
| (2.229) | (12.574) | (7.108) | (7.198) | |
| _cons | 63.333*** | |||
| (12.171) | ||||
| Observations | 11402 | 11402 | 11402 | 11402 |
| Pseudo R2 | .z | .z | .z | .z |
| Standard errors are in parentheses | ||||
| *** p<.01, ** p<.05, * p<.1 | ||||
My supervisor is asking for robustness check using quantile regression, however I have never worked with this before.
Best,
Peter
Comment