Fashion Regression Analysis

Regression Analysis

Dependent Variable: SALES |

Method: Least Squares |

|

Sample: 2 142 |

Included observations: 141 |

Variable | Coefficient | Std. Error | t-Statistic | Prob. |

C | 58049.10 | 1987.385 | 29.20879 | 0.0000 |

CCI | -265.0262 | 11249.30 | -0.023559 | 0.9812 |

EI | -17.88767 | 6746.690 | -0.002651 | 0.9979 |

PSI | 85.26887 | 4500.365 | 0.018947 | 0.9849 |

R-squared | 0.422220 | Mean dependent var | 39243.55 |

Adjusted R-squared | 0.409568 | S.D. dependent var | 6178.843 |

S.E. of regression | 4747.798 | Akaike info criterion | 19.79671 |

Sum squared resid | 3.09E+09 | Schwarz criterion | 19.88036 |

Log likelihood | -1391.668 | F-statistic | 33.37145 |

Durbin-Watson stat | 0.256155 | Prob(F-statistic) | 0.000000 |

Regression Equation:

Sales = 58049.10 – 265.03(consumer confidence index) – 17.89(expectations index) + 85.27(present situation index)

R2 Analysis

42.2% of total clothing sales are accounted for by the Consumer Confidence Index, Expectations Index, and Present Situation Index

T-stat Analysis

The P-values for the independent variables CCI and ECI are 0.22 and 0.48 respectively, indicating that the null hypotheses (H0: β1 = 0, H0: β3 = 0) cannot be rejected. Therefore the coefficients are not statistically significant. The P-value for ECI is 0.00, making it statistically significant on a 95% confidence interval. Although two of the coefficients are not significant the P-value for the regression as a whole is zero. This indicates that the entire regression is significant on a 95% confidence interval and that the null hypotheses are rejected while the alternate hypotheses (Ha: β1≠ 0, β2≠ 0, β3 ≠ 0) are accepted. The coefficients for CCI and ECI are not significant because they are correlated with CPI. A multicollinearity test was run in EViews and it was found that CCI and CPI are 73% correlated and ECI and CPI are 55% correlated. However, CCI and ECI are not correlated...