|
on Forecasting |
| By: | Tingting Cheng; Jiti Gao; Oliver Linton |
| Abstract: | We propose two new nonparametric predictive models: the multi-step nonparametric predictive regression model and the multi-step additive predictive regression model, in which the predictive variables are locally stationary time series. We define estimation methods and establish the large sample properties of these methods in the short horizon and the long horizon case. We apply our methods to stock return prediction using a number of standard predictors such as dividend yield. The empirical results show that all of these models can substantially outperform the traditional linear predictive regression model in terms of both in-sample and out-of-sample performance. In addition, we find that these models can always beat the historical mean model in terms of in-sample fitting, and also for some cases in terms of the out-of-sample forecasting. We also propose a trading strategy based on our methodology and show that it beats the buy and hold stategy provided the tuning parameters are well chosen. |
| Keywords: | Kernel estimator, locally stationary process, series estimator, stock return prediction |
| JEL: | C14 C22 G17 |
| Date: | 2019 |
| URL: | http://d.repec.org/n?u=RePEc:msh:ebswps:2019-4&r=all |
| By: | Novy Ann M. Etac; Roel F. Ceballos |
| Abstract: | This study was conducted to find an appropriate statistical model to forecast the volatilities of PSEi using the model Generalized Autoregressive Conditional Heteroskedasticity (GARCH). Using the R software, the log returns of PSEi is modeled using various ARIMA models and with the presence of heteroskedasticity, the log returns was modeled using GARCH. Based on the analysis, GARCH models are the most appropriate to use for the log returns of PSEi. Among the selected GARCH models, GARCH (1,2) has the lowest AIC value and also has the highest LL value implying that GARCH (1,2) is the best model for the log returns of PSEi. |
| Date: | 2019–02 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:1904.00749&r=all |
| By: | Gaglianone, Wagner; Giacomini, Raffaella; Issler, Joao; Skreta, Vasiliki |
| Abstract: | "Rational inattention" is becoming increasingly prominent in economic modelling, but there is little empirical evidence for its central premiseâ??that the choice of attention results from a cost-benefit optimization. Observational data typically do not allow researchers to infer attention choices from observables. We fill this gap in the literature by exploiting a unique dataset of professional forecasters who update their inflation forecasts at days of their choice. In the data we observe how many forecasters update (extensive margin of updating), the magnitude of the update (intensive margin), and the objective of optimiza- tion (forecast accuracy). There are also "shifters" in incentives: A contest that increases the benefit of accurate forecasting, and the release of official data that reduces the cost of information acquisition. These features allow us to link observables to attention and incentive parameters. We structurally estimate a model where the decision to update and the magnitude of the update are endogenous and the latter is the outcome of a rational inattention optimization. The model fits the data and gives realistic predictions. We find that shifts in incentives affect both extensive and intensive margins, but the shift in benefits from the contest has the largest aggregate effect. Counterfactuals reveal that accuracy is maximized if the contest coincides with the release of information, aligning higher benefits with lower costs of attention. |
| Keywords: | Contest; incentives; rational inattention; structural estimation; Survey Design |
| JEL: | D80 D83 E27 E37 |
| Date: | 2019–03 |
| URL: | http://d.repec.org/n?u=RePEc:cpr:ceprdp:13619&r=all |
| By: | Daniel Borup (Aarhus University and CREATES); Bent Jesper Christensen (Aarhus University and CREATES and the Dale T. Mortensen Center); Yunus Emre Ergemen (Aarhus University and CREATES) |
| Abstract: | This paper proposes tests of the null hypothesis that model-based forecasts are uninformative in panels, allowing for individual and interactive fixed effects that control for cross-sectional dependence, endogenous predictors, and both short-range and long-range dependence. We consider a Diebold-Mariano style test based on comparison of the model-based forecast and a nested nopredictability benchmark, an encompassing style test of the same null, and a test of pooled uninformativeness in the entire panel. A simulation study shows that the encompassing style test is reasonably sized in finite samples, whereas the Diebold-Mariano style test is oversized. Both tests have non-trivial local power. The methods are applied to the predictive relation between economic policy uncertainty and future stock market volatility in a multi-country analysis. |
| Keywords: | Panel data, predictability, long-range dependence, Diebold-Mariano test, encompassing test |
| JEL: | C12 C23 C33 C52 C53 |
| Date: | 2019–03–25 |
| URL: | http://d.repec.org/n?u=RePEc:aah:create:2019-04&r=all |
| By: | Pauwels, Laurent; Radchenko, Peter; Vasnev, Andrey |
| Abstract: | The majority of financial data exhibit asymmetry and heavy tails, which makes forecasting the entire density critically important. Recently, a forecast combina- tion methodology has been developed to combine predictive densities. We show that combining individual predictive densities that are skewed and/or heavy-tailed results in significantly reduced skewness and kurtosis. We propose a solution to over- come this problem by deriving optimal log score weights under Higher-order Moment Constraints (HMC). The statistical properties of these weights are investigated the- oretically and through a simulation study. Consistency and asymptotic distribution results for the optimal log score weights with and without high moment constraints are derived. An empirical application that uses the S&P 500 daily index returns illustrates that the proposed HMC weight density combinations perform very well relative to other combination methods. |
| Keywords: | Forecast combination; Predictive densities; Optimal weights; Skewness; Kurtosis |
| Date: | 2019–03–19 |
| URL: | http://d.repec.org/n?u=RePEc:syb:wpbsba:2123/20175&r=all |
| By: | Fokin, Nikita (Фокин, Никита) (The Russian Presidential Academy of National Economy and Public Administration) |
| Abstract: | This paper contains the construction of the large vector autoregression with 𝐿1 regularization on monthly data of Russian macroeconomic indicators taking into account the high dependence of the domestic economy on oil prices. The point of this work is to demonstrate the possibility and advantages of using the described approach to forecast Russian macroparameters using a large set of regressors, which from the theoretical point of view should improve the forecasts in comparison with models with a smaller dimension. Data on indices of industrial production, producer prices, investments, exports, imports, interest rates, indicators of consolidated and federal budgets, etc. were used. The final database consists of 45 variables with a total length of 15 years, for the period 2002M01-2016M12 - 180 points, from 44 regressors are endogenous, as well as one exogenous - the real oil price. The 𝐿1 regularization approach used in this paper allows us to estimate the model on such a large amount of data even if the observations are less than the number of estimated parameters. Based on the estimated model, we evaluate pseudo out of sample forecasts of indices of industrial production and the quality of the obtained forecasts was compared with the quality of the forecasts for the classical ARIMA model. The results of the evaluated model testify to the superiority of the evaluated model over all the benchmarks considered. |
| Keywords: | indices of industrial production, ARIMA model, VAR model, VAR-LASSO model, forecasting, impulse responses, long-run multipliers, oil prices |
| Date: | 2019–03 |
| URL: | http://d.repec.org/n?u=RePEc:rnp:wpaper:031951&r=all |
| By: | Manabu Asai (Faculty of Economics, Soka University, Japan); Rangan Gupta (Department of Economics, University of Pretoria, Pretoria, South Africa); Michael McAleer (Department of Finance, Asia University, Taiwan, Discipline of Business Analytics, University of Sydney Business School, Australia, Econometric Institute, Erasmus School of Economics, Erasmus University Rotterdam, The Netherlands, Department of Economic Analysis and ICAE, Complutense University of Madrid, Spain, and Institute of Advanced Sciences, Yokohama National University, Japan) |
| Abstract: | The paper investigates the impact of jumps in forecasting co-volatility in the presence of leverage effects. We modify the jump-robust covariance estimator of Koike (2016), such that the estimated matrix is positive definite. Using this approach, we can disentangle the estimates of the integrated co-volatility matrix and jump variations from the quadratic covariation matrix. Empirical results for daily crude oil and gold futures show that the co-jumps of the two futures have significant impacts on future co-volatility, but that the impact is negligible in forecasting weekly and monthly horizons. |
| Keywords: | Commodity Markets, Co-volatility, Forecasting, Jump, Leverage Effects, Realized Covariance, Threshold Estimation |
| JEL: | C32 C33 C58 Q02 |
| Date: | 2019–03 |
| URL: | http://d.repec.org/n?u=RePEc:pre:wpaper:201925&r=all |
| By: | Chan, Felix; Pauwels, Laurent |
| Abstract: | Forecasts are usually produced from models and expert judgements. The reconciliation of different forecasts presents an interesting challenge for managerial decisions. Mean absolute deviations and mean squared errors scoring rules are commonly employed as the criteria of optimality to aggregate or combine multiple forecasts into a consensus forecast. While much is known about mean squared errors in the context of forecast combination, little attention has been given to the mean absolute deviation. This paper establishes the first-order condition and the optimal solutions from minimizing mean absolute deviation. With this result, the paper derives the conditions in which the optimal solutions for minimizing mean absolute deviation and mean squared error loss functions are equivalent. More generally, this paper derives a sufficient condition which ensures the equivalence of optimal solutions of minimizing different loss functions under the same affine constraint that each feasible solution must sum to one. A simulation study and an illustration using expert forecasts data corroborate the theoretical findings. Interestingly, the numerical analysis shows that even with skewness in the data, the equivalence is unaffected. However, when outliers are presented in the data, mean absolute deviation is more robust than the mean squared error in small samples, which is consistent with the conventional belief relating the two loss functions. |
| Keywords: | Forecast combination; forecast accuracy; mean absolute deviation; optimal weights; |
| Date: | 2019–03–19 |
| URL: | http://d.repec.org/n?u=RePEc:syb:wpbsba:2123/20176&r=all |