Moving Average Penetration on a Monthly Timeframe
By
Rennie on Friday, February 19th, 2010 at 2:34 am
Cumulative TICK closed at +66,800 Thursday as NYSE TICK action has remained consistently positive for the past seven sessions, with no closing readings below +30,000. Note from the chart that the 20-day moving average is trending higher, and that’s likely to continue given that negative readings from late January will be falling off the average over the next seven sessions. A rising moving average usually means a rising market, so this should be a supportive factor.
The NYSE advance/decline ratio closed above 2.0 for the third day in a row Thursday and the fifth time in the last seven sessions. Note that the cumulative breadth line is closing in on the January highs as it continues to outperform the S&P, a generally positive sign.
The VIX fell for a seventh consecutive session Thursday, not something you see everyday given volatility’s mean-reverting nature. I found only 14 separate occurrences since 1993 in which the VIX posted seven consecutive down days…
VIX Down Seven Consecutive Sessions
02/18/10… S&P500 ??? one week later
11/10/09… S&P500 +1.6% one week later
10/13/09… S&P500 +1.7% one week later
04/10/07… S&P500 +1.6% one week later
08/17/06… S&P500 -0.1% one week later
05/24/05… S&P500 +0.7% one week later
12/23/04… S&P500 +0.2% one week later
11/10/04… S&P500 +1.6% one week later
04/15/03… S&P500 +3.2% one week later
10/26/01… S&P500 -1.6% one week later
11/23/98… S&P500 -1.1% one week later
02/04/98… S&P500 +1.3% one week later
11/21/97… S&P500 +1.2% one week later
06/17/93… S&P500 -0.4% one week later
01/18/93… S&P500 +0.7% one week later
In 10 out of 14 occurrences the S&P500 was higher one week later, suggesting the persistent drop in the VIX usually means volatility will stay subdued. I think the most interesting takeaway is that the S&P never gained or lost much more than 1.5% over the next week. With the S&P already off 1% in overnight trade, this suggests downside potential could be limited heading into next Thursday.
NYSE TICK posted its highest intraday low in over a month Thursday as institutional sellers were nowhere in sight, a pattern that has typically led to limited upside the following session. Apologies for not sending out an intraday update about this, it somehow went unnoticed until after the close. I’ll be updating my database over the weekend to ensure this type of miss doesn’t happen again. Here’s a look at the last thirty instances in which the NYSE TICK posted its highest low in three weeks and the S&P closed at a three-week high. Note only one case led to a gain of more than 0.5% the next session…
SPX 15-day High, NYSE TICK Highest Low in 15 Days
02/18/10… S&P500 ??? next day
12/24/09… S&P500 +0.1% next day
12/22/09… S&P500 +0.2% next day
11/09/09… S&P500 -0.0% next day
10/19/09… S&P500 -0.6% next day
09/22/09… S&P500 -1.0% next day
09/16/09… S&P500 -0.3% next day
08/21/09… S&P500 -0.1% next day
07/15/09… S&P500 +0.9% next day (*)
05/18/09… S&P500 -2.2% next day
04/09/09… S&P500 +0.3% next day
08/05/08… S&P500 +0.3% next day
05/15/08… S&P500 +0.2% next day
05/01/08… S&P500 +0.3% next day
05/30/07… S&P500 +0.0% next day
05/02/07… S&P500 +0.4% next day
04/05/07… S&P500 +0.1% next day
01/24/07… S&P500 -1.1% next day
10/16/06… S&P500 -0.4% next day
10/12/06… S&P500 +0.2% next day
10/04/06… S&P500 +0.2% next day
08/16/06… S&P500 +0.2% next day
07/03/06… S&P500 -0.7% next day
06/29/06… S&P500 -0.2% next day
05/05/06… S&P500 -0.1% next day
04/05/06… S&P500 -0.2% next day
02/16/06… S&P500 -0.2% next day
11/25/05… S&P500 -0.9% next day
11/02/05… S&P500 +0.4% next day
09/09/05… S&P500 -0.1% next day
08/02/05… S&P500 +0.1% next day
Stepping back for a longer-term perspective, I’ve recently noticed that standard moving averages on the monthly S&P chart can be unusually effective at signaling major shifts in the market with minimal whipsaws. See this chart of the last thirty years along with a 20-bar (month) moving average for an example. Buy arrows correspond with instances when the S&P500 initially closed above the 20-month average. Exits correspond with the first subsequent close back under the average. Below I’ve listed every buy & exit since 1950…
SPX Crosses Above/Below 20-Month Moving Average
11/30/09… Buy SPX 1095… Open
07/31/03… Buy SPX 990… 01/31/08 Exit 1378… +39%
01/31/95… Buy SPX 470… 11/30/00 Exit 1314… +180%
07/29/94… Buy SPX 458… 11/30/94 Exit 453… -1%
04/29/94… Buy SPX 450… 06/30/94 Exit 444… -1%
01/31/91… Buy SPX 343… 03/31/94 Exit 445… +30%
12/30/88… Buy SPX 277… 08/31/90 Exit 322… +16%
10/31/88… Buy SPX 278… 11/30/88 Exit 273… -2%
08/31/84… Buy SPX 166… 10/30/87 Exit 251… +51%
10/29/82… Buy SPX 133… 05/31/84 Exit 150… +13%
11/30/81… Buy SPX 126… 12/31/81 Exit 122… -3%
04/30/80… Buy SPX 106… 08/31/81 Exit 122… +15%
12/29/78… Buy SPX 96… 03/31/80 Exit 102… +6%
07/31/78… Buy SPX 100… 10/31/78 Exit 93… -7% (*)
04/30/75… Buy SPX 87… 03/31/77 Exit 98… +13%
12/31/70… Buy SPX 92… 04/30/73 Exit 106… +15%
01/31/67… Buy SPX 86… 06/30/69 Exit 97… +13%
03/29/63… Buy SPX 66… 05/31/66 Exit 86… +30%
01/31/63… Buy SPX 66… 02/28/63 Exit 64… -3%
12/30/60… Buy SPX 58… 05/31/62 Exit 59… +2%
06/30/60… Buy SPX 56… 07/29/60 Exit 55… -2%
05/29/58… Buy SPX 44… 03/31/60 Exit 55… +25%
04/30/57… Buy SPX 45… 08/30/57 Exit 45… 0%
01/29/54… Buy SPX 26… 01/31/56 Exit 44… +69%
09/30/49… Buy SPX 15… 05/29/53 Exit 24… +60%
Out of 24 signals since 1950, note that only one led to a meaningful loss when the signal was closed out, and even that was less than 10%. By comparison, there were fourteen signals that generated a profit of more than 10%. Since 1970, this simple timing strategy matched the performance of buy & hold while only in the market two-thirds of the time. Add the interest earned while in cash and it outperformed significantly.
The market’s consistent performance when buy signals are in effect suggests that a series of down months in the immediate future (or one big down month) that produces a break of the 20-month moving average is unlikely. Such a move would produce a loss in excess of 10% from the buy signal triggered last November (at roughly the same price as current levels), something that hasn’t occurred a single time in the last 60 years. If the recent buy is going to be a losing signal, a more likely scenario from the standpoint of this indicator is an extended period of sideways action that will enable the moving average to catch up with price by the end of the summer.
Moving Average Penetration on a Monthly Timeframe
By Rennie on Friday, February 19th, 2010 at 2:34 amCumulative TICK closed at +66,800 Thursday as NYSE TICK action has remained consistently positive for the past seven sessions, with no closing readings below +30,000. Note from the chart that the 20-day moving average is trending higher, and that’s likely to continue given that negative readings from late January will be falling off the average over the next seven sessions. A rising moving average usually means a rising market, so this should be a supportive factor.
The NYSE advance/decline ratio closed above 2.0 for the third day in a row Thursday and the fifth time in the last seven sessions. Note that the cumulative breadth line is closing in on the January highs as it continues to outperform the S&P, a generally positive sign.
The VIX fell for a seventh consecutive session Thursday, not something you see everyday given volatility’s mean-reverting nature. I found only 14 separate occurrences since 1993 in which the VIX posted seven consecutive down days…
VIX Down Seven Consecutive Sessions
02/18/10… S&P500 ??? one week later
11/10/09… S&P500 +1.6% one week later
10/13/09… S&P500 +1.7% one week later
04/10/07… S&P500 +1.6% one week later
08/17/06… S&P500 -0.1% one week later
05/24/05… S&P500 +0.7% one week later
12/23/04… S&P500 +0.2% one week later
11/10/04… S&P500 +1.6% one week later
04/15/03… S&P500 +3.2% one week later
10/26/01… S&P500 -1.6% one week later
11/23/98… S&P500 -1.1% one week later
02/04/98… S&P500 +1.3% one week later
11/21/97… S&P500 +1.2% one week later
06/17/93… S&P500 -0.4% one week later
01/18/93… S&P500 +0.7% one week later
In 10 out of 14 occurrences the S&P500 was higher one week later, suggesting the persistent drop in the VIX usually means volatility will stay subdued. I think the most interesting takeaway is that the S&P never gained or lost much more than 1.5% over the next week. With the S&P already off 1% in overnight trade, this suggests downside potential could be limited heading into next Thursday.
NYSE TICK posted its highest intraday low in over a month Thursday as institutional sellers were nowhere in sight, a pattern that has typically led to limited upside the following session. Apologies for not sending out an intraday update about this, it somehow went unnoticed until after the close. I’ll be updating my database over the weekend to ensure this type of miss doesn’t happen again. Here’s a look at the last thirty instances in which the NYSE TICK posted its highest low in three weeks and the S&P closed at a three-week high. Note only one case led to a gain of more than 0.5% the next session…
SPX 15-day High, NYSE TICK Highest Low in 15 Days
02/18/10… S&P500 ??? next day
12/24/09… S&P500 +0.1% next day
12/22/09… S&P500 +0.2% next day
11/09/09… S&P500 -0.0% next day
10/19/09… S&P500 -0.6% next day
09/22/09… S&P500 -1.0% next day
09/16/09… S&P500 -0.3% next day
08/21/09… S&P500 -0.1% next day
07/15/09… S&P500 +0.9% next day (*)
05/18/09… S&P500 -2.2% next day
04/09/09… S&P500 +0.3% next day
08/05/08… S&P500 +0.3% next day
05/15/08… S&P500 +0.2% next day
05/01/08… S&P500 +0.3% next day
05/30/07… S&P500 +0.0% next day
05/02/07… S&P500 +0.4% next day
04/05/07… S&P500 +0.1% next day
01/24/07… S&P500 -1.1% next day
10/16/06… S&P500 -0.4% next day
10/12/06… S&P500 +0.2% next day
10/04/06… S&P500 +0.2% next day
08/16/06… S&P500 +0.2% next day
07/03/06… S&P500 -0.7% next day
06/29/06… S&P500 -0.2% next day
05/05/06… S&P500 -0.1% next day
04/05/06… S&P500 -0.2% next day
02/16/06… S&P500 -0.2% next day
11/25/05… S&P500 -0.9% next day
11/02/05… S&P500 +0.4% next day
09/09/05… S&P500 -0.1% next day
08/02/05… S&P500 +0.1% next day
Stepping back for a longer-term perspective, I’ve recently noticed that standard moving averages on the monthly S&P chart can be unusually effective at signaling major shifts in the market with minimal whipsaws. See this chart of the last thirty years along with a 20-bar (month) moving average for an example. Buy arrows correspond with instances when the S&P500 initially closed above the 20-month average. Exits correspond with the first subsequent close back under the average. Below I’ve listed every buy & exit since 1950…
SPX Crosses Above/Below 20-Month Moving Average
11/30/09… Buy SPX 1095… Open
07/31/03… Buy SPX 990… 01/31/08 Exit 1378… +39%
01/31/95… Buy SPX 470… 11/30/00 Exit 1314… +180%
07/29/94… Buy SPX 458… 11/30/94 Exit 453… -1%
04/29/94… Buy SPX 450… 06/30/94 Exit 444… -1%
01/31/91… Buy SPX 343… 03/31/94 Exit 445… +30%
12/30/88… Buy SPX 277… 08/31/90 Exit 322… +16%
10/31/88… Buy SPX 278… 11/30/88 Exit 273… -2%
08/31/84… Buy SPX 166… 10/30/87 Exit 251… +51%
10/29/82… Buy SPX 133… 05/31/84 Exit 150… +13%
11/30/81… Buy SPX 126… 12/31/81 Exit 122… -3%
04/30/80… Buy SPX 106… 08/31/81 Exit 122… +15%
12/29/78… Buy SPX 96… 03/31/80 Exit 102… +6%
07/31/78… Buy SPX 100… 10/31/78 Exit 93… -7% (*)
04/30/75… Buy SPX 87… 03/31/77 Exit 98… +13%
12/31/70… Buy SPX 92… 04/30/73 Exit 106… +15%
01/31/67… Buy SPX 86… 06/30/69 Exit 97… +13%
03/29/63… Buy SPX 66… 05/31/66 Exit 86… +30%
01/31/63… Buy SPX 66… 02/28/63 Exit 64… -3%
12/30/60… Buy SPX 58… 05/31/62 Exit 59… +2%
06/30/60… Buy SPX 56… 07/29/60 Exit 55… -2%
05/29/58… Buy SPX 44… 03/31/60 Exit 55… +25%
04/30/57… Buy SPX 45… 08/30/57 Exit 45… 0%
01/29/54… Buy SPX 26… 01/31/56 Exit 44… +69%
09/30/49… Buy SPX 15… 05/29/53 Exit 24… +60%
Out of 24 signals since 1950, note that only one led to a meaningful loss when the signal was closed out, and even that was less than 10%. By comparison, there were fourteen signals that generated a profit of more than 10%. Since 1970, this simple timing strategy matched the performance of buy & hold while only in the market two-thirds of the time. Add the interest earned while in cash and it outperformed significantly.
The market’s consistent performance when buy signals are in effect suggests that a series of down months in the immediate future (or one big down month) that produces a break of the 20-month moving average is unlikely. Such a move would produce a loss in excess of 10% from the buy signal triggered last November (at roughly the same price as current levels), something that hasn’t occurred a single time in the last 60 years. If the recent buy is going to be a losing signal, a more likely scenario from the standpoint of this indicator is an extended period of sideways action that will enable the moving average to catch up with price by the end of the summer.