Eleven Time Series used in Gilbert (1995)
Data is for Canada unless otherwise indicated. The series start in February 1974 and end in September 1993 (236 observations on each variable).
The input series is 90 day interest rates (R90) and the ten output variables are CPI, GDP, M1, long run interest rates (RL), the Toronto stock exchange 300 index (TSE300), employment, the Canada/US exchange rate (PFX), a commodity price index in US dollars, US industrial production, and US CPI.
R90, RL and TSE are differenced. All other variables are in terms of percent change.
R90 is the 3 month prime corporate paper rate. While it is not set directly by the Bank of Canada, Bank policy influences it directly and it is often thought of as a proxy "policy variable."
The Statistics Canada identifiers are B14017 (R90), P484549 (CPI), I37026 (GDP), B1627 (M1), B14013 (RL), B4237 (TSE300), D767608 (employment), B3400 (PFX).
M.BCPI (commodity price index) is published by the Bank of Canada. JQIND (US industrial production), and CUSA0 (US CPI) are DRI identifiers.
The data for M1 (B1627) were taken prior to revisions made in December 1993.
This data is a TSdata object. The input series name is "R90" and the output series names are "CPI", "GDP", "M1", "RL", "TSE300", "employment", "PFX", "commod.price index", "US ind.prod." and "US CPI"
Statistics Canada, Bank of Canada, DRI.
Gilbert, P.D. 1995 "Combining VAR Estimation and State Space Model Reduction for Simple Good Predictions" J. of Forecasting: Special Issue on VAR Modelling. 14:229-250
Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.