The current rebalance settings allow you to select a 52 week rebalance schedule. One would assume this would be equivalent to a 1 year rebalance schedule. However, due to some funny date math it doesn’t actually behave that way.
For example, if I start a simulation on 2/1/1999 with a 52 week rebalance schedule, you might assume that the rebalance dates would be roughly as follows
2/1/1999
2/1/2000
2/1/2001
…
2/1/2019
Surprisingly this is not the case… instead the rebalance dates look as follows
2/1/1999
1/31/2000
1/29/2001
1/28/2002
…
1/22/2008
…
1/11/2016
…
1/07/2019
Now clearly this isn’t the end of the world, but is probably very different behavior than what most people would expect. It also kind of screws up my current attempt to analyze the seasonal effects of rebalancing a strategy in different months. (I expect I can find a workaround if I try.)
P123 staff were you aware of this phenomenon, and is this the behavior you intended the model to have?
Thanks,
Daniel