Skip to content

Relative ROC (RROC) as a Relative Stock Selection Mechanism (Part 1)

October 10, 2009

For all the work that is sometimes done on indicator refinement, sometimes its the simple things that work best. When your goal is to find the best stocks or ETFs, often indicators like RSI2 or DV2 are not completely sufficient as they do not account for relative volatility between stocks. That is, given two stocks with DV2 readings below 10—you could have widely different results because one could be a slow moving defensive stock, and the other could be a high tech stock.  The defensive stock could be down 3% on the week and the high tech stock could be down 12%. Other things being equal, to maximize return you will want to focus on the stock that has gone up or down the most–and the high tech stock will make you more money most of the time. This is a whole subject area in itself and differences in volatilities and betas, even alphas (the residual) have substantial impacts in the relative selection process.

Today we will look at a couple simple tests using only relative ROCs. I will use only S&P500 stocks to ensure that we do not include stocks that are untradeable. For this post we will focus only on 1-week returns (5-day ROC). Below are the returns to two different 20-stock portfolios: the stocks with the   20  best and worst five day ROCs (1-week returns) from the S&P500 index from December 28, 2007 to the beginning of October 2009. The strategy only involves buying the top or bottom 20 and exiting 1-week later and rebalancing . Thus there is no dynamic exit here, which understates the strategy performance.

RROC

12 Comments leave one →
  1. Damian permalink
    October 10, 2009 8:47 pm

    Is it really relative ROC, or just ROC? What’s relative about it?

    • david varadi permalink*
      October 10, 2009 9:38 pm

      hi damian its just ROC but i call it relative roc only because we aren’t looking at any absolute level of ROC, just ROC vs the universe. its just semantics. I don’t really want to call it relative strength only because it makes things a little confusing at short intervals–and i don’t want other people who are not familiar to confuse it with Relative Strength Index.

      cheers
      dv

  2. Bill permalink
    October 11, 2009 8:56 am

    David,

    do you use amibroker to code this, what would be the formula for percentrank in amibroker.
    Bill

    • October 11, 2009 9:07 pm

      Bill, no, just ROC(c, d) where d = however many days lookback you want to use. I think you’re the same Bill that I owe some AmiB code. I’m going to send right now.

  3. October 11, 2009 9:01 am

    David,

    What is the hedging strategy? Are you buying the best (worst) performers and then selling the worst (best) as the hedge? Or are you selling SPYs?

    Al

    • david varadi permalink*
      October 11, 2009 10:40 am

      hi, im buying the 20 worst and shorting the 20 best.
      cheers
      dv

  4. Damian permalink
    October 11, 2009 9:35 am

    What is the hedged strategy?

  5. Bill permalink
    October 11, 2009 10:30 am

    David,

    how did the hedged strategy perform before 2008

    Bill

  6. October 11, 2009 10:41 pm

    David, based on your day of the week seasonal study, do you think any edge could be added by an adaptive day of the week used to determine the days to open and close the positions?

    • david varadi permalink*
      October 12, 2009 1:35 am

      hi, i believe that the weekend effect could be a moderator across all stocks presuming one allowed dynamic exits. it is possible that the day of the week bias is present in certain stocks, however as presented, this effect weakened to the point of being untradeable in the last decade. I think personally the exit at open or close, and whether exit/entry occured around the weekends should matter most. of course week of month is actually a robust seasonal tendency based on my research so this would enhance things the most as you could still have the static 5 day exit, just enter during the most favorable week.

      dv

Trackbacks

  1. Blog Piggybacking: 5 Day Rate-of-Change | System Trading with Woodshedder

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: