Estimated coefficients of the ordinary least squares (OLS) and fixed effects (FE) regressions of relative income price (RIP) of bidis and smokeless tobacco in Bangladesh, 2009–2015
| OLS | FE | |||||
| Coefficient | Robust SE | P values | Coefficient | SE | P values | |
| Bidi | ||||||
| Trend | ||||||
| Wave 2 | −0.92 | 0.30 | 0.00 | −1.17 | 0.23 | 0.00 |
| Wave 3 | −1.85 | 0.26 | 0.00 | −1.86 | 0.24 | 0.00 |
| Wave 4 | −2.39 | 0.27 | 0.00 | −2.62 | 0.26 | 0.00 |
| Constant | 5.89 | 0.68 | 0.00 | 5.28 | 0.30 | 0.00 |
| Number of observations | 1603 | 1603 | ||||
| R2 | 0.24 | 0.10 | ||||
| Smokeless tobacco | ||||||
| Trend | ||||||
| Wave 4 | 0.33 | 0.63 | 0.60 | |||
| Constant | 4.45 | 3.22 | 0.17 | |||
| Number of observations | 399 | |||||
| R2 | 0.37 | |||||
The regression analysis controls for age, gender, education, occupation, socioeconomic status, residence and village fixed effects. The estimates for these control variables are suppressed for the brevity of presentation.
Reference categories include male, persons with 0 year of education, owner/tenant farmers, low socioeconomic status, urban area of residence and Wave 1 (2009).
The SEs of OLS estimates are adjusted for autocorrelation of error terms of multiple observations on the same individual.
The results of estimation for smokeless tobacco are based on data from International Tobacco Control (ITC) Bangladesh Survey Waves 3 and 4. The reference category for the trend variable is Wave 3.
The FE estimates for smokeless tobacco could not be obtained due to lack of sufficient number of individuals with repeated observations in Waves 3 and 4.