Identification of seasonality in time series: A note

Davey, Anne; Flores, Benito E.

doi:10.1016/0895-7177(93)90126-j

Cited by 19 publications

(11 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Time-series data require a careful investigation of any potential critical patterns before further analysis. The detection of possible seasonality in time-series data is important to draw conclusions on the proper methodology to use (Davey and Flores, 1993;Kapoor et al, 1981;Hanssens et al, 2002). In a first step, the time-series is visualised to gain a first impression of its composition (Figure 1).…”

Section: Methodsmentioning

confidence: 99%

Comparing the day temperature and holiday effects on retail sales of alcoholic beverages – a time-series analysis

Hirche

Haensch

Lockshin

2021

IJWBR

View full text Add to dashboard Cite

Purpose Little research on the influence of external factors, such as weather and holiday periods, on retail sales on alcoholic beverages is available. This study aims to investigate how weekly retail sales of different alcoholic beverages vary in association with daily maximum temperatures and annual federal holidays across selected US counties in the years 2013 to 2015. The research provides information, which can contribute to better sales forecasts. Design/methodology/approach Secondary data of weekly retail sales (volume) of alcoholic beverages from 37,346 stores in 651 counties in the USA are analysed. The data cover on average 21% of all existing US counties and 12% of the total US off-trade retail sales of alcoholic beverages in the period studied (Euromonitor, 2017). Additional data of federal holidays and meteorological data are collated for each county in the sample. Seasonal autoregressive integrated moving average models with exogenous regressors (SARIMAX) are applied to develop forecasting models and to investigate possible relationships and effects. Findings The results indicate that off-trade retail sales of beer, liquor, red and white wine are temperature sensitive throughout the year, while contrary to expectations rosé, sparkling and other wines are not. Sales sensitivities to temperature also differ by geography. In the warmest regions, liquor and white wine sales do not respond to temperature changes, as opposed to the coolest regions, where they are responsive. Public holidays, particularly Easter, Thanksgiving, Christmas and New Year holidays, represent a constant influencing factor on short-term sales increases for all investigated alcoholic beverage categories. Originality/value This is the first large-scale study of weather and holiday-related sales variations over time, across geographies and different alcoholic beverage categories. Seasonal and non-seasonal short-term sales variations are important for retailers and manufacturers alike. Accounting for expected changes in demand accommodates efficiencies along the supply chain and has implications for retail management, as well as adjusting marketing efforts in competing categories.

show abstract

Section: Methodsmentioning

confidence: 99%

Comparing the day temperature and holiday effects on retail sales of alcoholic beverages – a time-series analysis

Hirche

Haensch

Lockshin

2021

IJWBR

View full text Add to dashboard Cite

show abstract

“…If there is significant seasonality, the autocorrelation plot should show spikes at multiples of lags equal to the period, the seasonal lag (for quarterly data, we would expect to see significant spikes at lag 4,8,12,16, and so on). Davey and Flores [13] proposed a method which adds statistical tests of seasonal indexes for the multiplicative model that helps identify seasonality with greater confidence. Tests for seasonality are also given in Kendall and Ord [7].…”

Section: Introductionmentioning

confidence: 99%

Comparison of Two Time Series Decomposition Methods: Least Squares and Buys-Ballot Methods

Iwueze¹,

Nwogu²,

Nlebedim³

et al. 2016

OJS

View full text Add to dashboard Cite

This paper discusses comparison of two time series decomposition methods: The Least Squares Estimation (LSE) and Buys-Ballot Estimation (BBE) methods. As noted by Iwueze and Nwogu (2014), there exists a research gap for the choice of appropriate model for decomposition and detection of presence of seasonal effect in a series model. Estimates of trend parameters and seasonal indices are all that are needed to fill the research gap. However, these estimates are obtainable through the Least Squares Estimation (LSE) and Buys-Ballot Estimation (BBE) methods. Hence, there is need to compare estimates of the two methods and recommend. The comparison of the two methods is done using the Accuracy Measures (Mean Error (ME)), Mean Square Error (MSE), the Mean Absolute Error (MAE), and the Mean Absolute Percentage Error (MAPE). The results from simulated series show that for the additive model; the summary statistics (ME, MSE and MAE) for the two estimation methods and for all the selected trending curves are equal in all the simulations both in magnitude and direction. For the multiplicative model, results show that when a series is dominated by trend, the estimates of the parameters by both methods become less precise and differ more widely from each other. However, if conditions for successful transformation (using the logarithmic transform in linearizing the multiplicative model to additive model) are met, both of them give similar results.

show abstract

“…This idea was used by Hewitt, Milner, Csima, and Pakula (1971) but did not use a binary indicator as suggested by Edwards (1961), instead using all of the ranking information. Rogerson (1996) made an attempt to SOME TESTS FOR SEASONALITY IN TIME SERIES DATA 386 generalize this test by relaxing the relatively strict assumption of Hewitt et al (1971) that seasonality is only present if a six-month peak period is followed by a six-month trough period.…”

mentioning

confidence: 99%

“…Several statistics have also been proposed to test for seasonality. They can be broken down into three groups: the Chi-Square (χ 2 ) Goodness-of-Fit test and the Kolmogorov-Smirnov type statistic, the Harmonic analyses based on the Edwards' type statistic (Edwards, 1961), and the Nonparametric Tests.…”

mentioning

confidence: 99%