Quantopian's community platform is shutting down. Please read this post for more information and download your code.
Back to Community
A split cancellation not detected?

I found my algo goes sth wrong in backtesting recent days. By shorting the stock RKDA it would make me into debt horribly.

The research shows:

open_price high low close_price volume price
2018-01-22 00:00:00+00:00 0.28 0.318 0.260 0.300 1577373.0 0.300
2018-01-23 00:00:00+00:00 0.31 0.310 0.243 0.256 1217674.0 0.256
2018-01-24 00:00:00+00:00 5.28 7.020 4.650 5.356 329146.0 5.356
2018-01-25 00:00:00+00:00 5.34 5.930 4.860 5.410 123725.0 5.410

I noticed there was a 1/20 split on Jan 24 according to Yahoo Finance History:

Jan 25, 2018 5.34 5.93 4.86 5.41 5.41 142,400
Jan 24, 2018 5.28 7.02 4.65 5.38 5.38 368,600
Jan 23, 2018 1/20 Stock Split
Jan 23, 2018 6.20 6.20 4.86 5.12 5.12 69,600
Jan 22, 2018 5.60 6.36 5.20 6.00 6.00 90,600

But not shown on Nasdaq http://www.nasdaq.com/symbol/rkda/dividend-history

And the prices stay around $5 as usual for 2018 Jan, not around $0.2 as Q's after-adjusted records.

I wonder if there is a split canceled Q did not fix it. Deeply I doubt the result of backtests I have done for whether the profit came from dividend not accurately calculated.

1 response

I noticed the problematic data fixed.