Spurious deterministic seasonality and autocorrelation corrections with quarterly data: Further Monte Carlo results

Following recent work of Franses, Hylleberg and Lee (FHL), this paper analyses the consequences of fitting a deterministic seasonal model to a quarterly time series which can be (at least approximately) described by a seasonal unit root(s) model. Besides the distribution of the coefficient of determination, the empirical distributions of two commonly used statistics are also investigated through Monte Carlo experiments for small, moderately large and large samples. FHL's work is also extended allowing the possibility of residual autocorrelation corrections. The main conclusion that emerges from the results is that one should not try to measure the importance of deterministic seasonality nor test for its presence in the context of such (static) regression models, even when using some form of residual autocorrelation correction. A simple empirical application is provided to illustrate our results. First version received: July 1997/final version received: July 1998     

The January effect in the foreign exchange market: Evidence for seasonal equity carry trades

In this study, we investigate monthly seasonality in the foreign exchange market. Given the well-known recurrent higher returns in some month than in others in stock markets around the world, we consider it likely that a seasonal outperformance of a country’s stock market over another is associated with similar seasonal patterns in capital flows and exchange rates. A seasonal profit (carry trade) opportunity can be created by the simultaneous appreciation of a country’s currency and the outperformance of its stock market. By focusing on the world’s key currency pairs, the US dollar-Deutsche mark and the US dollar-euro, and by using a Markov-switching framework, we document persistent January and December effects in the foreign exchange market from 1971 to 2017. Analysis of the German-US stock returns differential and their bilateral capital flows reveal similar month effects in 65% of the whole sample.     

Buy or make? Agricultural production diversity,markets and dietary diversity in Afghanistan

Increasing on-farm production diversity and improving markets are recognized as ways to improve the dietary diversity of smallholders. Using instrumental variable methods to account for endogeneity, we study the interplay of production diversity, markets and diets in the context of seasonality in Afghanistan. We confirm an important seasonal dimension to the interplay. Improved crop diversity over the year is positively associated with dietary diversity in the regular season, but not in the lean season. Livestock species diversity remains important for dietary diversity throughout the year, but particularly so in the lean season when the influence of cropping diversity is low. Market aspects become important for dietary diversity specifically in the lean season.     

Yearly,monthly and weekly seasonality of tourism demand: A decomposition analysis

Annual seasonal variations in tourism demand have been a central theme in literature. However, annual seasonality is not the only time-based inequality in tourism flows that has important implications on policy-making decisions at destinations. Within the context of tourism, this study aims to make an in-depth analysis of intra-monthly and intra-weekly tourism demand using the entropy and relative redundancy measures as alternative seasonality indicators to the Gini coefficient in order to provide new tools to manage tourism and propose new action policies at these frequencies. In comparison with the Gini coefficient, the entropy measure is simpler to compute and it is easily decomposable. Using the case study of air arrivals and departures to and from the Balearic Islands, results show the appropriateness of entropy and relative redundancy as seasonal indicators but also as a new information tools for tourism seasonality analysis.     

The January effect across volatility regimes

Using a Markov regime switching model, this article presents evidence of the well-known January effect on stock returns. The specification allows a distinction to be drawn between two regimes: one with high volatility and another with low volatility. We obtain a time-varying January effect that is, in general, positive and significant in both volatility regimes. However, this effect is larger in the high-volatility regime. In sharp contrast with most of the previous literature, we find two major results: (1) the January effect exists for all sizes of portfolio; (2) the negative correlation between the magnitude of the January effect and portfolio size fails across volatility regimes. Moreover, our evidence supports a slight decline in the January effect for all sizes of portfolio except the smallest, for which it is even larger.     

Implications of climate change for tourism in Australia

This study assesses the impacts of projected climate change on Australia's tourism industry. Based on application of the Tourism Climatic Index, it investigates potential changes in climatic attractiveness for Australia's major destinations, and discusses implications for tourist flows and tourism planning, development and management. Australia may see considerable changes in spatial and temporal patterns of climatic suitability for tourism, with a southward shift in the most desirable conditions and a decline in the climatic attractiveness of northern locations. For destinations in which conditions are projected to decline, increased investment in indoor amenities may become increasingly necessary; where conditions are projected to improve, the provision of adequate infrastructure to accommodate potential increases in visitation and implementation of mechanisms to minimise the impacts of excess use may be more important. Adoption of a pro-active rather than reactive stance to climate change will maximise the ability of tourism stakeholders to successfully adapt.     

Sources of food price volatility and child malnutrition in Niger and Malawi

In recent times, considerable attention has been paid to the nutritional impact of the sharp hikes in the international food prices which took place in 2007–8 and 2010–11. While understandable, this growing focus has perhaps obscured the impact of other variables affecting malnutrition in Sub-Saharan Africa, i.e. the long-term impact of agricultural policies on food supply and prices, large and persistent seasonal variations in food prices, and the impact of famines which still affect parts of the continent. This paper focuses on the relative impact of these factors on child malnutrition (measured by the number of child admissions to feeding centres) in Malawi and Niger, two countries which closely represent the situation of other small, landlocked, subsistence agricultural economies facing severe food security problems. Our analysis shows that in these countries the drivers of changes in domestic staple prices and child malnutrition are related not only – or not primarily – to variations of international food prices but also to the impact of agricultural policies on food production and prices, in a persistent food price seasonality, and in recurrent and poorly managed famines. These factors can exert a strong upward pressure on food prices and child malnutrition even during years of falling international prices.     

Damping seasonal factors: Shrinkage estimators for the X-12-ARIMA program

We examine the effect of damping X-12-ARIMA's estimated seasonal variation on the accuracy of its seasonal adjustments of time series. Two methods for damping seasonals are proposed. In a simulation experiment, we generated time series data for each of 90 distinct experimental conditions that, in aggregate, characterize the variety of monthly series in the M3-competition. X-12-ARIMA consistently overestimated the actual seasonal variation by an amount consistent with statistical theory. Damping seasonals reduced X-12-ARIMA's estimation error by as much as 79% and under no conditions was estimation error increased beyond a trivial amount. Improvement depended primarily on the degree to which random variation in a series dominated seasonal variation. When the multiplicative X-12-ARIMA model did not match the data-generating model, overestimation was less for trend series than for series with no trend; otherwise the presence of trend had no discernible effect. One of the proposed methods was somewhat more accurate and robust, but more complex, than the other. In an analysis of real data—the 1428 monthly series of the M3-competition-damping X-12-ARIMA seasonals prior to forecasting (1) reduced the average forecasting MAPE by 4.9–1.4% and (2) improved forecasting accuracy for 59–65% of the series, depending on the forecasting horizon. This research suggests that damping X-12-ARIMA seasonals leads to more accurate seasonal adjustments of time series, thus providing a more reliable basis for policy-making, forecasting, and the evaluation of forecasting methods by researchers.     

Benchmarking by State Space Models

We have a monthly series of observations which are obtained from sample surveys and are therefore subject to survey errors. We also have a series of annual values, called benchmarks, which are either exact or are substantially more accurate than the survey observations; these can be either annual totals or accurate values of the underlying variable at a particular month. The benchmarking problem is the problem of adjusting the monthly series to be consistent with the annual values. We provide two solutions to this problem. The first of these is a two-stage method in which we first fit a state space model to the monthly data alone and then combine the results obtained at this stage with the benchmark data. In the second solution we construct a single series from the monthly and annual values together and fit a state space model to this series in a single stage. The treatment is extended to series which behave multiplicatively. The methods are illustrated by applying them to Canadian retail sales sereis.     

Time Series Modelling of Daily Tax Revenues

We provide a detailed discussion of time series modelling of daily data in general and daily tax revenues in particular. The main feature of the daily tax revenue series is the pattern within calendar months. Standard time series methods for seasonal adjustment and forecasting cannot be used since the number of banking days per calendar month varies and because there are two levels of seasonality: between months and within months. We propose a daily time series model based on unobserved components that allows for the classic decomposition into trend, seasonal plus irregular, but it also includes components for intra-monthly, trading-day and length-of-month effects. Such components typically rely on stochastic cubic spline, polynomial and dummy variable functions. State space techniques are used for the recursive computation of the likelihood and forecasts functions with special allowance for irregular spacing. The model is operational for daily forecasting at the Dutch Ministry of Finance. We present the model specification and discuss estimation and forecasting results up to December 1999. A comparative forecast evaluation is also presented.