Cointegrated time series associated with financial and sports gambling markets are analyzed in terms of time-varying parameter models. Parameter drift is modeled in terms of lagged disequilibria. Model forecasts are intended to capitalize on periods of market inefficiency. Modeling premises are that (1) present and past disequilibria—shocks both within and between time series—may affect subsequent changes and rates of these changes within individual series and (2) sufficiently large shocks may disrupt/alter model structure such that resulting forecasts may be temporarily unreliable. Reduced forecasting equations are in terms of higher order ARMA models that are not limited to bilinear processes. Sports forecasting models based on public information are usually more effective—in terms of profitable trading/wagering strategies—than those for the financial sector for two reasons: (1) Insider information is less prevalent. (2) Modeling is simplified since lagged shocks associated with the gambling lines/spreads are known—in contrast with financial modeling where there are no comparable gambling shocks, only unknown, lagged statistical shocks in terms of MA variables. Forecasting is illustrated for NFL and NBA playoff games. In financial markets, cointegration is discussed in terms of candlestick chart variants with modeling illustrations given in terms of recent Google price changes.
Download Info
To download:
If you experience problems downloading a file, check if you have the
proper application to
view it first. Information about this may be contained
in the File-Format links below. In case of further problems read
the IDEAS help
page. Note that these files are not on the IDEAS
site. Please be patient as the files may be large.
Volume (Year): 2 (2008) Issue (Month): 3 (December) Pages: 15-32 Download reference. The following formats are available: HTML
(with abstract),
plain text
(with abstract),
BibTeX,
RIS (EndNote, RefMan, ProCite),
ReDIF