|
on Forecasting |
| By: | Ali Dib; Mohamed Gammoudi; Kevin Moran |
| Abstract: | The authors document the out-of-sample forecasting accuracy of the New Keynesian model for Canada. They estimate their variant of the model on a series of rolling subsamples, computing out-of-sample forecasts one to eight quarters ahead at each step. They compare these forecasts with those arising from simple vector autoregression (VAR) models, using econometric tests of forecasting accuracy. Their results show that the forecasting accuracy of the New Keynesian model compares favourably with that of the benchmarks, particularly as the forecasting horizon increases. These results suggest that the model could become a useful forecasting tool for Canadian time series. The authors invoke the principle of parsimony to explain their findings. |
| Keywords: | Business fluctuations and cycles; Economic models; Econometric and statistical methods |
| JEL: | E32 E37 C12 |
| Date: | 2006 |
| URL: | http://d.repec.org/n?u=RePEc:bca:bocawp:06-4&r=for |
| By: | Marc-André Gosselin; René Lalonde |
| Abstract: | The analysis and forecasting of developments in the U.S. economy have always played a critical role in the formulation of Canadian economic and financial policy. Thus, the Bank places considerable importance on generating internal forecasts of U.S. economic activity as an input to the Canadian projection. Over the past year, Bank staff have been using a new macroeconometric model, MUSE (Model of the U.S. Economy). The model is a system of estimated equations that describe, in a stock-flow framework, the interactions among the principal macroeconomic variables, such as gross domestic product (GDP), inflation, interest rates, and the exchange rate. The stock-flow equilibrium is fully described in MUSE. In steady state, the model defines specific values for all stocks, including capital stock, government debt, financial wealth, and net foreign assets. In MUSE, most behavioural equations are governed by a polynomial adjustment cost (PAC) structure. This approach is widely used in the U.S. Federal Reserve Board's FRB/US model. By allowing for lags in the dynamic equations in the context of forward-looking rational expectations, the PAC approach strikes a balance between theoretical structure and forecasting accuracy. MUSE, therefore, makes an explicit distinction between dynamic movements caused by changes in expectations and those caused by adjustment costs. Moreover, GDP is decomposed into household expenditures, business investment, government spending, exports, and imports. Hence, MUSE can be used to predict the consequences of a wide variety of shocks to the U.S. economy. |
| Keywords: | Economic models; Business fluctuations and cycles |
| JEL: | E37 C53 E17 E27 F17 |
| Date: | 2005 |
| URL: | http://d.repec.org/n?u=RePEc:bca:bocatr:96&r=for |
| By: | Zagaglia, Paolo (Dept. of Economics, Stockholm University) |
| Abstract: | Yes, but only at short horizons from 1 to 3 quarters over the full post-World War II sample. The predictive relation between the yield spread and the output gap is characterized by parameter instability. Differently from the predictive models of the yield spread for output growth, structural instability is not due to a loss of predictive ability after 1985. Rather, the predictive relation estimated on post-1985 data holds for a range of horizons larger than for pre-1985 data. I also show that the information on current monetary policy is statistically irrelevant for the prediction of the output gap over the post-1985 subsample. |
| Keywords: | output gap; yield spread; predictability |
| JEL: | E27 E43 |
| Date: | 2006–05–08 |
| URL: | http://d.repec.org/n?u=RePEc:hhs:sunrpe:2006_0005&r=for |
| By: | Zagaglia, Paolo (Dept. of Economics, Stockholm University) |
| Abstract: | I apply a multiresolution decomposition to the term spread and real-GDP growth in the U.S. Using the filtered data, I study whether the yield spread helps forecasting output. The results show that the predictive power of the yield spread varies largely across time scales both in-sample and out-of-sample at various forecast horizons. Contrarily to the existing literature, I find evidence of a strikingly negative long-run relationship between the spread and future GDP growth over a frequency that spans from 8 to 16 years per cycle. A linear combination among filtered yield spreads shows a sizable improvement in forecasting out-of-sample. The decomposed series are also used for proposing a solution to the breakdown in the in-sample predictive relationship documented by Dotsey (1998) that occurs after 1985. |
| Keywords: | wavelets; term structure; predictability |
| JEL: | C19 E27 E43 |
| Date: | 2006–06–16 |
| URL: | http://d.repec.org/n?u=RePEc:hhs:sunrpe:2006_0004&r=for |
| By: | Francis X. Diebold (Department of Economics, University of Pennsylvania); Lei Ji (Department of Economics, University of Pennsylvania); Canlin Li (Graduate School of Management, University of California) |
| Abstract: | We assess and apply the term-structure model introduced by Nelson and Siegel (1987) and re-interpreted by Diebold and Li (2003) as a modern three-factor model of level, slope and curvature. First, we ask whether the model is a member of the affine class, and we find that it is not. Hence the poor forecasting performance recently documented for affine term structure models in no way implies that our model will forecast poorly, which is consistent with Diebold and Li's (2003) finding that it indeed forecasts quite well. Next, having clarified the relationship between our three-factor model and the affine class, we proceed to assess its adequacy directly, by testing whether its level, slope and curvature factors do indeed capture systematic risk. We find that they do, and that they are therefore priced. Finally, confident in the ability of our three-factor model to capture the pricing relations present in the data, we proceed to explore its efficacy in bond portfolio risk management. Traditional Macaulay duration is appropriate only in a one-factor (level) context; hence we move to a three-factor generalized duration, and we show the superior performance of hedges constructed using it. |
| Keywords: | Term structure; Yield curve; Factor model; Risk Management |
| JEL: | G1 E43 E47 C5 |
| Date: | 2006–03–09 |
| URL: | http://d.repec.org/n?u=RePEc:pen:papers:06-017&r=for |
| By: | Boriss Siliverstovs; Konstantin A. Kholodilin |
| Abstract: | This paper suggests a novel approach to pre-selection of the component series of the diffusion index based on their individual forecasting performance. It is shown that this targeted selection allows substantially improving the forecasting ability compared to the diffusion index models that are based on the largest available dataset. |
| Keywords: | Diffusion index, forecasting, optimal subset of data |
| JEL: | E32 C10 |
| Date: | 2006 |
| URL: | http://d.repec.org/n?u=RePEc:diw:diwwpp:dp598&r=for |
| By: | Jean-Thomas Bernard; Lynda Khalaf; Maral Kichian; Sebastien McMahon |
| Abstract: | Fluctuations in the prices of various natural resource products are of concern in both policy and business circles; hence, it is important to develop accurate price forecasts. Structural models provide valuable insights into the causes of price movements, but they are not necessarily the best suited for forecasting given the multiplicity of known and unknown factors that affect supply and demand conditions in these markets. Parsimonious representations of price processes often prove more useful for forecasting purposes. Central questions in such stochastic models often revolve around the time-varying trend, the stochastic convenience yield and volatility, and mean reversion. The authors seek to assess and compare alternative approaches to modelling these effects, focusing on forecast performance. Three econometric specifications are considered that cover the most up-to-date models in the recent literature on commodity prices: (i) random-walk models with autoregressive conditional heteroscedasticity (ARCH) or generalized ARCH (GARCH) effects, and with normal or student-t innovations, (ii) Poisson-based jump-diffusion models with ARCH or GARCH effects, and with normal or student-t innovations, and (iii) meanreverting models that allow for uncertainty in equilibrium price. |
| Keywords: | Econometric and statistical methods |
| JEL: | C52 C53 E37 |
| Date: | 2006 |
| URL: | http://d.repec.org/n?u=RePEc:bca:bocawp:06-14&r=for |
| By: | Jamie Armour |
| Abstract: | The author provides a statistical evaluation of various measures of core inflation for Canada. The criteria used to evaluate the measures are lack of bias, low variability relative to total CPI inflation, and ability to forecast actual and trend total CPI inflation. The author uses the same methodology as Hogan, Johnson, and Laflèche (2001) and thus provides updated empirical results. The findings are that most traditional measures of core inflation are unbiased and all continue to be less volatile than total inflation. They nevertheless display some volatility and have limited predictive ability. Overall, CPIW seems to have a slight advantage over the other measures, but the differences across measures are not large. (CPIW uses all components of total CPI but adjusts the weight of each component by a factor that is inversely proportional to the component's variability.) Compared with the results of Hogan, Johnson, and Laflèche, CPIW's relative performance has improved. The distribution of price changes for 54 CPI subcomponents is also examined, and substantial increases in both the skewness and kurtosis of this distribution since 1998 are found. |
| Keywords: | Inflation and prices |
| JEL: | E31 |
| Date: | 2006 |
| URL: | http://d.repec.org/n?u=RePEc:bca:bocawp:06-10&r=for |