which model works best for this data? Explain. Model 1: OLS, using observations

Question

which model works best for this data? Explain.

Model 1: OLS, using observations 1-158 Dependent variable: cost Coefficient Std. Error tratiop-value -4.74698 2.1610 0.00554157 0.000117403 47.20 0.0001 ** -2.197 0.0295* const Mean dependent var Sum squared resid R-squared F(1, 156) Log-likelihood Schwarz criterion 53.26996 S.D. dependent var 7786743 S.?. of regression 0.934563 Adjusted R-squared 2227.960 P-value(F) 87.05933 22.34167 0.934143 2.84e-94 1432.011 1434.499 -714.0056 Akaike criterion 1438.136 Hannan-Quinn TC-4.75+01Q Model 2: OLS, using observations 1-158 Dependent variable: cost Coefficient Std Error ratiop-value 0.9677 0.00434763 0.000230941 18.83 1.63160e-08 2.79915e-09 5.829 0.3347 0.0001 0.0001 2.22486 2.29906 const 9 53.26996 S.D. dependent var 63867.60 S.E. of regression 0.946328 Adjusted R-squared 1366.447 P-value F) 87.05933 20.29899 0.945635 3.59e-99 1402.696 1406.427 Mean dependent var Sum squared resid squared E(2, 155) Log-likelihood Schwarz criterion -698.3481 Akaike criterion 1411.884 Hannan-Quinn TC 2.22 + .00 X Q + 1.63 X Q-2

Explanation / Answer

Model 2 is working best for this data set. This is because coefficients of output and output squared are significant while using model 2. The p-values are also low and R squared is around 94 per cent which states that 94 per cent of the deviations in the cost variable are explained by output and output squared. The joint hypothesis testing using F test is also significant as F value is very high. Thus, model 2 best fits the data among all the three models.

Related Questions

Navigate

• Browse (All)
• Browse W
• Subjects

---

---

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.