First algorithm here =). Feedbacks are welcomed. I was wondering if it would be a better idea to long stocks with promising signs and short the reverse instead of the current implementation. Thanks

Thanks for sharing Dan! This looks interesting - it might be worth running the backtest over a longer period of time to see more history.

Also, if you're considering live trading the algorithm, I would put your global variables into context in initialize(). For example,

def initialize(context):  
  context.COND1_USED = True  
  context.SMA_RATIO_LOWER_BOUND = 1.0  
  context.COND2_USED = True  

Context is used to save state of variables between bars of the backtest and in live trading. Nice first algo!

Hi Dan look sharp, thanks for sharing. I was reading through the code and I don;t quite understand a few bits. If you don't mind, could you through some text around what rerank does and how it impacts the trade execution?
Cheers

Hi Rick, thank you for looking at my code. The rerank basically is called once everyday ATM, it calls other function to rerank the stocks based on their momentum. The momentum is calculated by looking at two MAVG, one is 2 days the other is 1 day. The rerank function makes sure that everyday you're buying the stock that has a positive momentum, and sell the stocks that have a negative momentum. Thanks

Hi Alisa, did you guys make changes to the MAVG function. Because right now if I deploy my code minutely like this

def initialize(context):
pass

def handle_data(context, data):
order(sid(24), 50)
log.info(str(data[sid(41575)].mavg(1)))
log.info(str(data[sid(41575)].mavg(4)))

The mavg(1) and mavg(4) are returning the same result. Any idea why this is happening? Thanks.

Hi Dan, This behavior is now fixed, go ahead and try it again. Thanks!