- Tradestation Labs provided a nice study to analyze jobless claims data, and subsequent movement in the following 5 minutes, and on the day.
- Volume increased on average 530% in the 5 minutes following the report release compared to the average of the previous three 5 minute bars.
- The correlation to the movement before the release and 5 minutes after the release is 0.66, implying a modest bias to continue in the direction of the prior movement.
- The correlation falls after the first 5 minutes, implying that it is not profitable to necessarily go with the directional movement of the first 5 minutes after the release.
- However, for worse than expected results, some of the correlation for the next 5 minutes is maintained, so “the market may be more likely to sell off when there is a worse-than-expected jobless claims announcement than to rally after a better-than-expected jobless claims announcement during the first five minutes of trading following the release.”
- There is no correlation to the 4:15 pm close, so the data does not impact the rest of the day.
- Source : TradestationLabs (requires tradesation ID)
- HistorySquared
- The tradestation code and accompanying spreadsheet could be adapted to analyze all manner of reports and instruments while adding in numerous other conditions, such as whether the underlying instrument was overbought or oversold going into the report – creating a buy the rumour sell the news scenario.
Popularity: 3% [?]