Application of statistics - Essay Example

Question 6 Regression on women regress sal age yos edyrs hf bf Source | SS df MS Number of obs = 1116 F( 5, 1110) = 176.27Model| 2.1080e+115 4.2159e+10Prob > F= 0.0000Residual| 2.6548e+111110 239170791R-squared= 0.4426Adj R-squared= 0.4401Total| 4.7628e+111115 427153397Root MSE= 15465sal| Coef.Std. Err. tP>|t|[95% Conf.Interval]age| -11.1776960.19331 -0.190.853-129.2832106.9278yos| 862.366962.2805 13.850.000740.1661984.5677edyrs| 5157.439217.1515 23.750.0004731.3665583.513hf| -8408.8642719.968 -3.090.002-13745.72-3072.006bf| -5001.221290.443 -3.880.000-7533.202-2469.237_cons| -35147.723984.731 -8.820.000-42966.17-27329.26 The predicted regression equation is sal. = -35147.72 -11.

17769 Age + 5157.439 edyrs – 8408.864 hf – 5001.22 bf and it is statistically significant at least 99% level since the p-value is 0.000. Therefore, for each salary increase the age of women decrease by 11.17769 while education years increase by 5157.439. The regression output also indicates that R2 is 0.4426 meaning that the independent variables can explain 44.26% of the variability of the salary of women. Regression on menSource|SSdf MSNumber of obs= 1116F( 5, 1110)= 168.87Model|2.0577e+115 4.

1154e+10Prob > F= 0.0000Residual|2.7051e+111110 243700320R-squared= 0.4320Adj R-squared= 0.4295Total|4.7628e+111115 427153397Root MSE= 15611sal|Coef.Std. Err. tP>|t|[95% Conf.Interval]age|22.3680560.41665 0.370.711-96.17567140.9118yos|853.113762.73202 13.600.000730.027976.2004edyrs|5346.696213.5179 25.040.0004927.7525765.641bm|-1720.0072057.91 -0.840.403-5757.842317.825hm|2870.3122999.173 0.960.339-3014.3758755_cons|-40354.593863.33 -10.450.000-47934.85-32774.34 The predicted regression equation is sal. = -40354.59 + 2853.

1137 Age + 5346.696 edyrs – 1720.007 hm+ 2870.312 hm – 40354.59 bm and it is statistically significant at least 99% level since the p-value is 0.000. Therefore, for each salary increase the age of women increase by 2853.1137 while education years increase by 5346.696. The regression output also indicates that R2 is 0.4320 meaning that the independent variables can explain 43.20 % of the variability of the salary of men.Question 7 and 8Source|SSdf MSNumber of obs= 1116F( 9, 1106)= 105.93Model|2.

2049e+119 2.4499e+10Prob > F= 0.0000Residual|2.5579e+111106 231271445R-squared= 0.4629Adj R-squared= 0.4586Total|4.7628e+111115 427153397Root MSE= 15208sal|Coef.Std. Err. tP>|t|[95% Conf.Interval]blckmale|-5651.82078.818 -2.720.007-9730.672-1572.928maleasian|3898.6413260.35 1.200.232-2498.52810295.81malehispanic|-877.36252976.633 -0.290.768-6717.8484963.123maleedyrs|-177.4859263.0927 -0.670.500-693.703338.7311maleyos|-144.9187122.9871 -1.180.239-386.23396.39562maleage|267.1671102.294 2.610.00966.45484467.8793age|-169.851381.14831 -2.090.037-329.0733-10.62929yos|951.403490.82714 10.470.000773.19051129.616edyrs|4989.166252.1224 19.790.0004494.4745483.858bm|0(omitted)hm|0(omitted)_cons|-31459.963934.081 -8.000.000-39179.06-23740.86 The predicted regression equation is sal.

= -31459 -169.8513 Age + 498.166 edyrs + 0 hm+ 0 bm-5651.8 blackmale +3898.644 maleasian – 877.3625 malehispanic – 177.4859 maleedyrs -144.9187maleyos + 267.1677maleage + 951.4034yos and it is statistically significant at least 99% level since the p-value is 0.000. Therefore, for each salary increase the age of women reduces by 31459 while bm and bh the relationship is zero..