Market timing using combined forecasts and machine learning

Successful market timing strategies depend on superior forecasting ability. We use a sentiment index model, a kitchen sink logistic regression model, and a machine learning model (least absolute shrinkage and selection operator, LASSO) to forecast 1‐month‐ahead S&P 500 Index returns. In order to determine how successful each strategy is at forecasting the market direction, a “beta optimization” strategy is implemented. We find that the LASSO model outperforms the other models with consistently higher annual returns and lower monthly drawdowns.     

Monthly Beta Forecasting with Low‐, Medium‐ and High‐Frequency Stock Returns

This paper evaluates the accuracy of 1‐month‐ahead systematic (beta) risk forecasts in three return measurement settings; monthly, daily and 30 minutes. It was found that the popular Fama–MacBeth beta from 5 years of monthly returns generates the most accurate beta forecast among estimators based on monthly returns. A realized beta estimator from daily returns over the prior year generates the most accurate beta forecast among estimators based on daily returns. A realized beta estimator from 30‐minute returns over the prior 2 months generates the most accurate beta forecast among estimators based on 30‐minute returns. In environments where low‐, medium‐ and high‐frequency returns are accurately available, beta forecasting with low‐frequency returns are the least accurate and beta forecasting with high‐frequency returns are the most accurate. The improvements in precision of the beta forecasts are demonstrated in portfolio optimization for a targeted beta exposure. Copyright © 2016 John Wiley & Sons, Ltd.     

Conditional predictability of daily exchange rates

At what forecast horizon is one time series more predictable than another? This paper applies the Diebold–Kilian conditional predictability measure to assess the out‐of‐sample performance of three alternative models of daily GBP/USD and DEM/USD exchange rate returns. Predictability is defined as a non‐linear statistic of a model's relative expected losses at short and long forecast horizons, allowing flexible choice of both the estimation procedure and loss function. The long horizon is set to 2 weeks and one month ahead and forecasts evaluated according to MSE loss. Bootstrap methodology is used to estimate the data's conditional predictability using GARCH models. This is then compared to predictability under a random walk and a model using the prediction bias in uncovered interest parity (UIP). We find that both exchange rates are less predictable using GARCH than using a random walk, but they are more predictable using UIP than a random walk. Predictability using GARCH is relatively higher for the 2‐weeks‐than for the 1‐month long forecast horizon. Comparing the results using a random walk to that using UIP reveals ‘pockets’ of predictability, that is, particular short horizons for which predictability using the random walk exceeds that using UIP, or vice versa. Overall, GBP/USD returns appear more predictable than DEM/USD returns at short horizons. Copyright © 2002 John Wiley & Sons, Ltd.     

Forecasting Daily Variations of Stock Index Returns with a Multifractal Model of Realized Volatility

Multifractal models have recently been introduced as a new type of data‐generating process for asset returns and other financial data. Here we propose an adaptation of this model for realized volatility. We estimate this new model via generalized method of moments and perform forecasting by means of best linear forecasts derived via the Levinson–Durbin algorithm. Its out‐of‐sample performance is compared against other popular time series specifications. Using an intra‐day dataset for five major international stock market indices, we find that the the multifractal model for realized volatility improves upon forecasts of its earlier counterparts based on daily returns and of many other volatility models. While the more traditional RV‐ARFIMA model comes out as the most successful model (in terms of the number of cases in which it has the best forecasts for all combinations of forecast horizons and evaluation criteria), the new model performs often significantly better during the turbulent times of the recent financial crisis. Copyright © 2014 John Wiley & Sons, Ltd.     

Forecasting and trading strategies based on a price trend model

In this paper, we consider the price trend model in which it is assumed that the time series of a security's prices contain a stochastic trend component which remains constant on each of a sequence of time intervals, with each interval having random duration. A quasi‐maximum likelihood method is used to estimate the model parameters. Optimal one‐step‐ahead forecasts of returns are derived. The trading rule based on these forecasts is constructed and is found to bear similarity to a popular trading rule based on moving averages. When applying the methods to forecast the returns of the Hang Seng Index Futures in Hong Kong, we find that the performance of the newly developed trading rule is satisfactory. Copyright © 2000 John Wiley & Sons, Ltd.     

Forecasting house prices in OECD economies

In this paper, we forecast real house price growth of 16 OECD countries using information from domestic macroeconomic indicators and global measures of the housing market. Consistent with the findings for the US housing market, we find that the forecasts from an autoregressive model dominate the forecasts from the random walk model for most of the countries in our sample. More importantly, we find that the forecasts from a bivariate model that includes economically important domestic macroeconomic variables and two global indicators of the housing market significantly improve upon the univariate autoregressive model forecasts. Among all the variables, the mean square forecast error from the model with the country's domestic interest rates has the best performance for most of the countries. The country's income, industrial production, and stock markets are also found to have valuable information about the future movements in real house price growth. There is also some evidence supporting the influence of the global housing price growth in out‐of‐sample forecasting of real house price growth in these OECD countries.     

Flow of conjunctural information and forecast of euro area economic activity

Combining forecasts, we analyse the role of information flow in computing short‐term forecasts up to one quarter ahead for the euro area GDP and its main components. A dataset of 114 monthly indicators is set up and simple bridge equations are estimated. The individual forecasts are then pooled, using different weighting schemes. To take into consideration the release calendar of each indicator, six forecasts are compiled successively during the quarter. We found that the sequencing of information determines the weight allocated to each block of indicators, especially when the first month of hard data becomes available. This conclusion extends the findings of the recent literature. Moreover, when combining forecasts, two weighting schemes are found to outperform the equal weighting scheme in almost all cases. Compared to an AR forecast, these improve by more than 40% the forecast performance for GDP in the current and next quarter. Copyright © 2010 John Wiley & Sons, Ltd.     

Are 16‐month‐ahead forecasts useful? A directional analysis of Japanese GDP forecasts

Past literature casts doubt on the ability of long‐term macroeconomic forecasts to predict the direction of change. We re‐examine this issue using the Japanese GDP forecast data of 37 institutions, and find that their 16‐month‐ahead forecasts contain valuable information on whether the growth rate accelerates or not. Copyright © 2006 John Wiley _ Sons, Ltd.     

Forecasting US Recessions with a Large Set of Predictors

In this paper, I use a large set of macroeconomic and financial predictors to forecast US recession periods. I adopt Bayesian methodology with shrinkage in the parameters of the probit model for the binary time series tracking the state of the economy. The in‐sample and out‐of‐sample results show that utilizing a large cross‐section of indicators yields superior US recession forecasts in comparison to a number of parsimonious benchmark models. Moreover, the data‐rich probit model gives similar accuracy to the factor‐based model for the 1‐month‐ahead forecasts, while it provides superior performance for 1‐year‐ahead predictions. Finally, in a pseudo‐real‐time application for the Great Recession, I find that the large probit model with shrinkage is able to pick up the recession signals in a timely fashion and does well in comparison to the more parsimonious specification and to nonparametric alternatives. Copyright © 2016 John Wiley & Sons, Ltd.     

Forecasting the Treasury's balance at the Fed

As part of the Fed's daily operating procedure, the Federal Reserve Bank of New York, the Board of Governors and the Treasury make a forecast of that day's Treasury balance at the Fed. These forecasts are an integral part of the Fed's daily operating procedure. Errors in these forecasts can generate variation in reserve supply and, consequently, the federal funds rate. This paper evaluates the accuracy of these forecasts. The evidence suggests that each agency's forecast contributes to the optimal, i.e., minimum variance, forecast and that the Trading Desk of the Federal Reserve Bank of New York incorporates information from all three of the agency forecasts in conducting daily open market operations. Moreover, these forecasts encompass the forecast of an economic model. Copyright © 2004 John Wiley & Sons, Ltd.     

Predicting returns and volatility with macroeconomic variables: evidence from tests of encompassing

Tests of forecast encompassing are used to evaluate one‐step‐ahead forecasts of S&P Composite index returns and volatility. It is found that forecasts over the 1990s made from models that include macroeconomic variables tend to be encompassed by those made from a benchmark model which does not include macroeconomic variables. However, macroeconomic variables are found to add significant information to forecasts of returns and volatility over the 1970s. Often in empirical research on forecasting stock index returns and volatility, in‐sample information criteria are used to rank potential forecasting models. Here, none of the forecasting models for the 1970s that include macroeconomic variables are, on the basis of information criteria, preferred to the relevant benchmark specification. Thus, had investors used information criteria to choose between the models used for forecasting over the 1970s considered in this paper, the predictability that tests of encompassing reveal would not have been exploited. Copyright © 2005 John Wiley & Sons, Ltd.     

Identification of asymmetric prediction intervals through causal forces

When causal forces are specified, the expected direction of the trend can be compared with the trend based on extrapolation. Series in which the expected trend conflicts with the extrapolated trend are called contrary series. We hypothesized that contrary series would have asymmetric forecast errors, with larger errors in the direction of the expected trend. Using annual series that contained minimal information about causality, we examined 671 contrary forecasts. As expected, most (81%) of the errors were in the direction of the causal forces. Also as expected, the asymmetries were more likely for longer forecast horizons; for six‐year‐ahead forecasts, 89% of the forecasts were in the expected direction. The asymmetries were often substantial. Contrary series should be flagged and treated separately when prediction intervals are estimated, perhaps by shifting the interval in the direction of the causal forces. Copyright © 2001 John Wiley & Sons, Ltd.     

Causal forces: Structuring knowledge for time-series extrapolation

This paper examines a strategy for structuring one type of domain knowledge for use in extrapolation. It does so by representing information about causality and using this domain knowledge to select and combine forecasts. We use five categories to express causal impacts upon trends: growth, decay, supporting, opposing, and regressing. An identification of causal forces aided in the determination of weights for combining extrapolation forecasts. These weights improved average ex ante forecast accuracy when tested on 104 annual economic and demographic time series. Gains in accuracy were greatest when (1) the causal forces were clearly specified and (2) stronger causal effects were expected, as in longer-range forecasts. One rule suggested by this analysis was: ‘Do not extrapolate trends if they are contrary to causal forces.’ We tested this rule by comparing forecasts from a method that implicitly assumes supporting trends (Holt's exponential smoothing) with forecasts from the random walk. Use of the rule improved accuracy for 20 series where the trends were contrary; the MdAPE (Median Absolute Percentage Error) was 18% less for the random walk on 20 one-year ahead forecasts and 40% less for 20 six-year-ahead forecasts. We then applied the rule to four other data sets. Here, the MdAPE for the random walk forecasts was 17% less than Holt's error for 943 short-range forecasts and 43% less for 723 long-range forecasts. Our study suggests that the causal assumptions implicit in traditional extrapolation methods are inappropriate for many applications.     

Predicting Stock Return Volatility: Can We Benefit from Regression Models for Return Intervals?

We study the performance of recently developed linear regression models for interval data when it comes to forecasting the uncertainty surrounding future stock returns. These interval data models use easy‐to‐compute daily return intervals during the modeling, estimation and forecasting stage. They have to stand up to comparable point‐data models of the well‐known capital asset pricing model type—which employ single daily returns based on successive closing prices and might allow for GARCH effects—in a comprehensive out‐of‐sample forecasting competition. The latter comprises roughly 1000 daily observations on all 30 stocks that constitute the DAX, Germany's main stock index, for a period covering both the calm market phase before and the more turbulent times during the recent financial crisis. The interval data models clearly outperform simple random walk benchmarks as well as the point‐data competitors in the great majority of cases. This result does not only hold when one‐day‐ahead forecasts of the conditional variance are considered, but is even more evident when the focus is on forecasting the width or the exact location of the next day's return interval. Regression models based on interval arithmetic thus prove to be a promising alternative to established point‐data volatility forecasting tools. Copyright ©2015 John Wiley & Sons, Ltd.     

Short‐term salmon price forecasting

This study establishes a benchmark for short‐term salmon price forecasting. The weekly spot price of Norwegian farmed Atlantic salmon is predicted 1–5 weeks ahead using data from 2007 to 2014. Sixteen alternative forecasting methods are considered, ranging from classical time series models to customized machine learning techniques to salmon futures prices. The best predictions are delivered by k‐nearest neighbors method for 1 week ahead; vector error correction model estimated using elastic net regularization for 2 and 3 weeks ahead; and futures prices for 4 and 5 weeks ahead. While the nominal gains in forecast accuracy over a naïve benchmark are small, the economic value of the forecasts is considerable. Using a simple trading strategy for timing the sales based on price forecasts could increase the net profit of a salmon farmer by around 7%.     

Forecasting the 10‐year US Treasury rate

This study compares the performance of two forecasting models of the 10‐year Treasury rate: a random walk (RW) model and an augmented‐autoregressive (A‐A) model which utilizes the information in the expected inflation rate. For 1993–2008, the RW and A‐A forecasts (with different lead times and forecast horizons) are generally unbiased and accurately predict directional change under symmetric loss. However, the A‐A forecasts outperform the RW, suggesting that the expected inflation rate (as a leading indicator) helps improve forecast accuracy. This finding is important since bond market efficiency implies that the RW forecasts are optimal and cannot be improved. Copyright © 2009 John Wiley & Sons, Ltd.     

Can We Predict the Financial Markets Based on Google's Search Queries?

We look into the interaction of Google's search queries and several aspects of international equity markets. Using a novel methodology for selecting words and a vector autoregressive modeling approach, we study whether the search queries of finance‐related words can have an impact on returns, volatility of returns and traded volume in four different English‐speaking countries. We identify several words whose search frequency is associated with changes in the dependent variables. In particular, we find that increases in search queries including the word stock predict increased volatility and decreased index returns over the next week. On top of that, we investigate the performance of a market‐timing strategy based on the search frequency of this word and benchmark it against random words from the Word‐Net database and a naive buy‐and‐hold strategy. The results of this empirical application are positive and particularly stronger during the global crisis of 2009. Copyright © 2016 John Wiley & Sons, Ltd.     

Economic factors and the stock market: a new perspective

It has been widely accepted that many financial and economic variables are non‐linear, and neural networks can model flexible linear or non‐linear relationships among variables. The present paper deals with an important issue: Can the many studies in the finance literature evidencing predictability of stock returns by means of linear regression be improved by a neural network? We show that the predictive accuracy can be improved by a neural network, and the results largely hold out‐of‐sample. Both the neural network and linear forecasts show significant market timing ability. While the switching portfolio based on the linear forecasts outperforms the buy‐and‐hold market portfolio under all three transaction cost scenarios, the switching portfolio based on the neural network forecasts beats the market only if there is no transaction cost. Copyright © 1999 John Wiley & Sons, Ltd.     

Forecasting volatility with outliers in GARCH models

In this paper, we detect and correct abnormal returns in 17 French stocks returns and the French index CAC40 from additive‐outlier detection method in GARCH models developed by Franses and Ghijsels (1999) and extended to innovative outliers by Charles and Darné (2005). We study the effects of outlying observations on several popular econometric tests. Moreover, we show that the parameters of the equation governing the volatility dynamics are biased when we do not take into account additive and innovative outliers. Finally, we show that the volatility forecast is better when the data are cleaned of outliers for several step‐ahead forecasts (short, medium‐ and long‐term) even if we consider a GARCH‐t process. Copyright © 2008 John Wiley & Sons, Ltd.