So my student have a question that i cannot answer as well. She analyzing the effect of ICT toward labor productivity using 8 years data panel using 4 independent variables with EVIEWS 13. Frankly i quite surprised that the R-squared value on her results is 0.94 with only 2 significance variables. Theoretically, in simple regression model with higher value of R-squared most likely indicates bad and have statistics problems. Recently, i asked her to calculate the data using STATA and the results shows that only have 0,51 R-Square with exact similar coefficient.
I've search some articles about it and says that eviews might be wrong, and some says that STATA is wrong. Can someone explain what should i do and which software have to use?
note:
1. Some articles says to using areg command in stata to find similar value as eviews, but i quite doubt because areg is using for categorical regression in stata and its not quite fit in panel regression model.
2. Some says that eviews is wrong calculation.
This is an amazing example! Fixed effects (within) regression and panel least squares (or pooled OLS) regressions are inherently different due to their treatment of individual heterogeneity and how they treat variance, hence the R2 values from these models are not directly comparable!
In short, the fixed effects focuses on within-level variance (i.e., how well the model explains the variation in the dependent variable after accounting for individual-specific means) while panel least squares captures both within-and between-individual variation, meaning this model doesn't control for individual-specific effects and you should expect to see wildly different R2
to several reasons:
1.Rounding Differences: STATA and Eviews may use different rounding conventions
2.Calculation Methodology: use slightly different algorithms or formulas to calculate the R-squared value. While the underlying concept is the same, the specific implementation can result in minor numerical differences.
3. Missing Data: may have different approaches to handling missing data in the dataset, which can impact the R-squared calculation.
Or Degrees of Freedom maydifferent approaches to this adjustment.
these differences are typically very small and may not have a significant impact.
- Badges
- Science topic
- Similar topics
- Geoscience
- Cartography
- Mapping