RDP 2010-03: Modelling Inflation in Australia Appendix B: Additional Regression Results

David Norman and Anthony Richards

June 2010

Table B1: Additional Regression Results
Standard Phillips curve equations
πe UR ΔUR Δmp πt−k Δ4oil Δ4M3 Adj-R2 AIC LR-value(c)
Baseline 0.403*** 0.142*** −0.003*** 0.079***       0.648 −9.91  
With πt−k   0.052** −0.003*** 0.040 0.166***     0.576 −9.73  
Consumer πe 0.715*** 0.047* −0.001 0.039       0.519 −9.60  
Oil price model 0.431*** 0.129*** −0.003*** 0.069**   0.009   0.640 −9.86  
Money model(a) 0.270*** −0.069 −0.005*** 0.071**     4.508 0.636 −9.87 3.86**
Mark-up model equations
πe Gap Δulc Δmp πt−k Δ4oil Δ4M3 Adj-R2 AIC
Baseline 0.309*** 0.178*** 0.172* 0.107***       0.618 −9.81  
With πt−k   0.057 0.209* 0.097** 0.107**     0.580 −9.72  
Consumer πe 0.444*** 0.030 0.254*** 0.106***       0.614 −9.80  
Vertical restriction 0.244(b) 0.167** 0.484*** 0.268***       0.480 −9.50  
Oil price model 0.340*** 0.141** 0.185** 0.091***   0.018**   0.638 −9.83 7.61*
New-Keynesian Phillips curve (using OLS)
πe Gap rulc rmp Δrmp πt−k real oil Adj-R2 AIC
Baseline 0.361*** 0.166*** 0.125** 0.027** −0.008 0.148   0.585 −9.75  
Oil price model 0.382*** 0.163*** 0.103** 0.054** −0.010 0.096 0.001 0.591 −9.75  
Notes: ***, ** and * represent significance at the 1, 5 and 10 per cent levels respectively. Where multiple lags included, coefficients shown are sum of the lags. Coefficients on output gap, real unit labour costs, real import prices and OECD output gap multiplied by 4 to convert to annual effect; coefficients on four-quarter change in M3 and oil prices divided by 4 to convert to elasticity.
(a) Estimated from 1990:Q1 to 2007:Q1. Over that sample, the baseline model has an adjustment R-squared of 0.622 and an AIC of −9.84.
(b) No standard errors are applicable for this coefficient.
(c) Chi-squared statistic from likelihood ratio test that the baseline (restricted) model is significantly different.