Sunday, July 12, 2009
Saturday, July 11, 2009
PREDICTING THE DAILY DOW JONES CLOSE USING GOOGLE TRENDS
Until the arrival of Google and other search engines, access to data that reflects the economic activities of massive numbers of consumers has been utterly inadequate. In the past we have had to rely on biased surveys, inadequate data collections, and the whims of consumer responses to our efforts to generate clear, precise, correct, and inclusive information needed to serve as the foundation for forcasting economic trends.
Interesting information is rapidly accumulating indicating that Google trend data may have finally provided us with a tool to forecast a host of economic trends by simply searching for the appropriate key phrases www.google.com/trends. A recent paper published by Hal Varian, Chief Economist and Hyunyoung Choi, Decision Support Engineering Analyst at Google outlines the theoretical basis for this claim: "Our work to date is summarized in a paper called Predicting the Present with Google Trends. We find that Google Trends data can help improve forecasts of the current level of activity for a number of different economic time series, including automobile sales, home sales, retail sales, and travel behavior."
I propose that there is no greater indicator of economic activity than that manifested by fluctuations in the UP or DOWN close of the Dow Jones. I further propose that the trend concept applies to ever decreasing incremental time frames: years, months, weeks, days, and even hours. I have developed a statistical model utilizing daily trend changes in key words that can predict ahead of time rather the Dow Jones will close UP or DOWN at the end of the trading day.
I WILL POST DAILY BY 8:00AM THE DAY THE MARKET OPENS THE RESULTS INDICATING RATHER THE DOW JONES WILL CLOSE UP OR DOWN ON THAT TRADING DAY.
Interesting information is rapidly accumulating indicating that Google trend data may have finally provided us with a tool to forecast a host of economic trends by simply searching for the appropriate key phrases www.google.com/trends. A recent paper published by Hal Varian, Chief Economist and Hyunyoung Choi, Decision Support Engineering Analyst at Google outlines the theoretical basis for this claim: "Our work to date is summarized in a paper called Predicting the Present with Google Trends. We find that Google Trends data can help improve forecasts of the current level of activity for a number of different economic time series, including automobile sales, home sales, retail sales, and travel behavior."
I propose that there is no greater indicator of economic activity than that manifested by fluctuations in the UP or DOWN close of the Dow Jones. I further propose that the trend concept applies to ever decreasing incremental time frames: years, months, weeks, days, and even hours. I have developed a statistical model utilizing daily trend changes in key words that can predict ahead of time rather the Dow Jones will close UP or DOWN at the end of the trading day.
I WILL POST DAILY BY 8:00AM THE DAY THE MARKET OPENS THE RESULTS INDICATING RATHER THE DOW JONES WILL CLOSE UP OR DOWN ON THAT TRADING DAY.
Subscribe to:
Posts (Atom)
Posts
