# Using VXO to time the SP 500, part 2

Happy New Year!

This is a follow-up to the popular post I did last January on using weekly closes in VXO to time the SP 500. In that post I showed it has historically been much more profitable to own the SP 500 following weeks when VXO rose vs. weeks when VXO declined. Beautifully simple.

In this post I’ll add another indicator to the timing model. Just a bit more complexity. This indicator will show that many weeks following a VXO decline are really not that bad. We can separate those that have tended to decline from those that tended to rise.

The indicator I’ll introduce is an acceleration measure based on SP 500 weekly highs and weekly closes. First I calculate each week’s high divided by the previous week’s close. The final indicator just looks at whether this number increases or decreases from week to week.

In other words, is HIGH t/CLOSE t-1 greater than or less than HIGH t-1/CLOSE t-2 ? I’ll refer to the high divided by previous close as H/C in the rest of the post.

In the chart below, you’ll see three equity curves. One invests in the SP 500 whenever VXO declines the prior week. The other two will distinguish weeks with an accelerating H/C vs. weeks with a decelerating H/C.

Nice. As you can see, the index tends to fall when H/C accelerates and rise when H/C decelerates. We are also closer to having a decent short setup based on times when VXO declines and H/C accelerates. We can make the short trade more appealing by simply requiring the previous week’s VXO close be above 19. A VXO close of 19 works well but other values around it work well too.

The below chart shows profit curves for investing in the index when VXO declines, H/C accelerates and either VXO closes above or below 19 the prior week.

Finally, let’s combine all long and short trades together. Here is how the final model rules could work.

**Long SP 500** when either VXO rose **or** when BOTH VXO declined and H/C decelerated.

**Short SP 500** when VXO declined, H/C accelerated, and VXO closed **greater** than 19.

**Cash** when VXO declined, H/C accelerated, and VXO closed **less** than 19.

Below are equity curves for buy and hold, a long only model, and a long/short model. Not bad.

Cheers and best of luck to you in 2014!

-Mike

****Technical note**: All data are from Yahoo finance and exclude dividends, commissions, fees, slippage, etc. etc.

_______________

**Update**: Here is an Excel file with data and calculations for the last chart. Enjoy! SPY VXO Weekly Strat Part 2

_______________

**Update2**: Here are some stats that were requested.

Posted on January 2, 2014, in Other. Bookmark the permalink. 33 Comments.

Great stuff, Mike! You’re using the intraday S&P 500 weekly high and not closes for your calculation, correct?

Thanks Joel! Yes that is correct about intraday highs and not simply the highest close of the week.

Thanks, Mike.

Could you share your thoughts on the robustness of this system? Do you think that it has a good chance of working in the future? Would you invest in it yourself?

And, how is fatherhood?

Alex

Hey Alex, good to hear from you.

Fatherhood has been a fun and exhausting experience. Makes me appreciate what my own parents did for me as a baby.

I think the system in this post is pretty good. I’ll be tracking it going forward and will watch to see that the predicted worst days continue to underperform. That’s really what this system is about – being invested most of the time and either avoiding or shorting when odds are the worst. I may post model allocations/ performance to the site. Haven’t decided yet.

Nice article Mike! I always enjoy your posts.

Thank you!

Two more things, Mike.

1. I would love for you to track this system on your blog.

2. Why does your equity curve stop at the beginning of 2012? I’m curious about what the system has done since, especially with VXO being rather flat.

And, I wish you and your expanding family a wonderful New Year.

It actually goes nearly through the end of 2013 (just missing the last week since I pulled the data earlier). It’s just the last label that stops in 2012. The system picked up most of the index gains in 2013.

Which is max drawdown of the strategy (L+S)? I get a mdd of -32%, do you agree?

Yup, we match on that.

One more question, if you don’t mind. Are the strategy’s calculations based on selling, buying, or shorting SPY at Friday close of a given week (say the first week being tested) or at the open of the following Monday?

I tested both trading at Friday’s close and trading at Monday’s open (the post uses Friday’s close). Makes almost no difference. That means you could see the signal after Friday’s close and put in your order for Monday’s open at your leisure.

I tested using the SP 500 index. For actual execution, you could use SPY or similar product. I would still use the index for all signals though that shouldn’t matter much either.

Excellent work! Do you have the Excel File on how to program this as well? Nice Job!

Thanks John, I should be able to post an Excel file sometime this weekend. Check back in a day or two.

Thanks Mike!

Mike, Are the results much different if you use VIX instead of VXO?

Good question. I remember from my research last year that VIX didn’t work as good. Will have to revisit that and also try out the new indicator with it. Haven’t done that yet.

Thanks for Posting Mike! Quick question. Could you explain to me how this calculation plays out? I am new to excel and trying to learn from your work.

=IF(AND(E1460J1459,E1460>19),((I1460-I1461)/I1460+1)*L1460,IF(AND(E1460J1459,E1460<19),L1460,I1461/I1460*L1460))

Seems like the equation is a bit garbled in your reply. I can walk you through it. I’ll use the formula in cell L5 as the lower row number makes it easier to read.

“=IF(AND(E4J3,E4>19),((I4-I5)/I4+1)*L4″

The equation starts out with an IF statement. Using AND() requires all conditions (separated by commas) be met for the following calculation ((I4-I5)/I4+1)*L4 to calculate. Ok, what does this do? The conditions inside the first AND() are checking to see if our short criteria are met (VXO fell in prior week, H/C rose in prior week from week ago, VXO closed above 19 in prior week). If all these conditions are met, then I use the formula that follows to calculate the return of a short trade and move the index forward by this percentage.

“,IF(AND(E4J3,E4=19 or <=19 in the formula). Otherwise it's possible to signal a long trade when we'd rather be in cash or short.

And wordpress managed to garble my reply as well. That’s bad. Feel free to email me at dontfearthebear at gmail dot com if you still want help.

Good strategy. Is there any added benefit to trading this strategy on a shorter timeframe from weekly to daily or even 2 or 3 day time periods?

Thanks. I have not tested on shorter time periods but would be a good idea to try.

That would be great if you could. It could possibly avoid some of those larger drawdowns. Just an FYI, even though the 3 x leveraged S&P 500 ETFs didnt exist in 1986, if they did and you invested in those instead, the $100k account would be worth about $12 billion today if you could stomach through some of the larger drawdowns as in 2008.

Initial testing doesn’t look good for higher frequency signals/trading. I guess this is consistent with what MarketSci found after I published my original post last January. The last day of the week works best and calculating signals on other days doesn’t add value from what I’ve tried. Disappointing but it is what it is.

I can still llive with the $12 billion version. Thanks for trying though. MarketSci had some good stuff on DVI indicator with SPX too. Wonder if that would work as an added dimension to this strategy? Will let you know if I come up with any other ideas.

Mike, Can you post stats on this system such as avg win, avg loss, avg gain and loss, percentages etc? Thanks! Also, have you gone long for this coming week?

When I get some time, I’ll run some numbers for you.

The system is long this week.

Thanks very much Mike. Are you planning on trading with it?

I’m hesitant to trade this since performance is lacking if we use other days of the week to calculate the signals. That is a red flag to me. I should have tested that earlier on.

At the least, I think the short setup is a good warning indicator of weak upcoming performance. I’ll give this strat some time to develop out of sample results. I leave it to the reader to examine the data and come to their own conclusions.

Mike, are you using Monday’s open or the previous weeks close to determine if VXO rose or fell for the week? Thanks.

Hi Joel, using previous week’s close.

Hello Mike, I am interested in determining whether the SP500 is in “follow through” or “mean reversion” mode. Would you have a suggestion for what tools to use to make such a determination? Have you ever investigated the SP500, or has your focus been confined to volatility XVI? Thanks much for sharing your work.

Hello Mike, Jim P. asked a good question about tools to use to make a determination “follow through” or “mean reversion” mode.

Great work