Article / 05 May 2014 at 14:14 GMT

A deep correction's on the horizon; how prepared are you?

Blogger / MoreLiver's Daily
Finland

• China PMI falls for fourth consecutive month
• ECB shows little sign of action on inflation
• Ukraine crisis still a major factor

Monday’s fall in European stock markets means the week has started on a sour note. The most important bad news came from China, where Markit reported that the manufacturing purchasing manager’s index (PMI) had shown contraction for the fourth month in a row (see press release: Business conditions continue to deteriorate in April). That dropped the AUDJPY spot below its recent range, and after recovering to pre-PMI levels, the latest Sentix investor confidence was published and being clearly below consensus forecast, the markets turned down again. The historical relationship between the Sentix and stock market is illustrated by Mads Koefoed’s graph.

In addition, it seems that traders expect that the European Central Bank (ECB) will fail to do anything at Thursday’s policy-setting meeting. The central bank is waiting for the new ECB staff inflation forecasts to be published, pushing the earliest date of monetary policy changes to June. As the European Commission today published its spring 2014 economic forecasts, which revised the 2014 inflation forecast down from 1.0 to 0.8 percent and the 2015 forecast from 1.3 percent to 1.2 percent, the ECB's reluctance to act just shows how empty its toolbox really is.

The crisis in Ukraine is also not showing any signs of going away quickly, which means uncertainty will continue, investments will be postponed and the economic sanctions will be seen to dent growth. Is this just a bad start to the week, or a sign of darker times ahead? I think the stars are aligning to suggest an increased risk of a market correction in the coming months. 

The arguments for a deeper correction are piling up

Two estimable commentators are seeing trouble ahead for the stock markets. Their main arguments are based on sentiment, market breadth, bond markets, stretched valuation, seasonals, and relative sector performance.

Cam Hui posted on his blog with the headline The bearish verdict from market cycle analysis, where he looked at the different sectors of the stock market, and noted that the relative performance of the sectors suggests an increased risk of a correction ahead. He has previously written about the deteriorating market internals, seasonality and the underappreciated macro risks caused by China and the Ukrainian crisis. 

Urban Carmel’s Weekly Market Summary notes that the market breadth divergence and negative seasonal effects are happening simultaneously, putting the market in a similar situation as in July 2011 or April 2012. The period from May to October tends to exhibit flat returns, and during the past 10 years there has been a notable market correction in every year except one.

Sell in May and go away?

For the time-constrained, one chart tells the story. For the past 20 years, an investor who has been long the S&P 500 only during the periods from May to September, has earned flat returns. All the stock market gains have happened during the periods from October to April.

While the May-September periods have not, on average, exhibited negative price changes, the markets have often gone through deep corrections. This has made the risk-reward ratio for holding stocks negative, especially for the weak hands that might not be able to remain invested during a deeper correction. 

Wesley L. Gray recently ran a test with different asset classes, including bonds and factors such as value and momentum. The results confirmed the May-October effect for most assets, and concluded that the effect is especially strong for stocks. The tendency of other factors to exhibit the same pattern as stocks is not that surprising, if one remembers that during times of flat returns or market corrections many institutional investors have had to lighten up their positions. This tendency is one of the reasons why cross-asset correlations tend to increase during market routs. For more on the "Sell in May", see Lance Roberts.

AUDJPY thermometer somewhat negative

In my article last week, Why AUDJPY is the Swiss army knife for traders I wrote about the strong relationship between the AUDJPY crossrate and the equity markets. On February 3, all three markets turned higher together but after the next low on March 14, the stock markets failed to convincingly follow the AUDJPY’s surge to new highs. While the DAX-index move higher was relatively large, it failed to make new highs, and the whole move higher could be seen as a reaction to the previous slump caused by the Ukrainian crisis. Ukraine-related headlines weigh more on the German market than the US.

The downward-sloping trendline in DAX since the January’s highs has held, and the market has lost momentum. The S&P 500 has been essentially range-trading since early March. At the same time, AUDJPY has broken below mid-April’s trading range, and looks as if it could be ready to break below its recent trading range as well. 

For the moment, it looks as if the resistance level in DAX will hold and cap the markets, especially if the AUDJPY does not begin to exhibit signs of strength. The next five months could very well see volatile and flat equity markets, including a nasty correction of roughly 10 percent. With Europe’s crisis countries enjoying exceptionally low bond yields and many betting on the ECB to save the day, I believe it is very important to adjust one’s investments to be able to tolerate a larger correction.

See also Teis Knuthsen’s excellent Investment Outlook: Stocks on the rise but only just.

AUDJPY daily

Source: Saxo Trader

AUDJPY hourly
















Source: Saxo Trader


Helsinki-based Juhani Huopainen blogs at MoreLiver's Daily. His specialities include options (also exotic), the euro crisis and EURUSD.

 

5y
Mickette Mickette
Juhani, always good to read your comments!
5y
Mickette Mickette
Juhani, do you believe the US dollar is going to retest the 79.00 level in the very short term?
5y
DP DP
Wow, I'm impressed by "Sell in May" chart analysis!
5y
Juhani Huopainen Juhani Huopainen
DP: I hope you were referring to the charts of the two articles I referred to :) I merely intended to update the AUDJPY / equity charts here, without any particularly new insights - I put them in writing on the "Swiss Army Knife"-article.

Mickette: 79.00-level? USD versus what?
5y
DP DP
Juhani -- I referred to "CHART OF THE DAY: 'Sell In May And Go Away'" by Sam Ro linked by you in the current article. It is very convincing.

What can we expect from, say, WTI crude in case of deep indices correction, same deep plunge, I guess?
5y
Juhani Huopainen Juhani Huopainen
The correlation between oil and stock index returns is non-existent, see http://www.clevelandfed.org/research/trends/2008/0908/04ecoact.cfm
That does not mean that there is no statistical relationship between the two - they might still be cointegrated - i.e. levels might matter, and they might move together with a lag, or the correlation might be in place only at certain times. Perhaps oil reacts more to real economy demand, oil supply shocks (or risks of shocks), while S&P is more open to financial economy drivers?

A chart of S&P 500 and continues WTI futures suggests that the bottoms in S&P 500 do not meaningfully align with bottoms in oil
5y
DP DP
Very interesting and thanks for the effort to make the chart.

I am agreeing that it's possible to imagine the situation when S&P plunges, but oil demand is sort of "unaffected", so there is no correlation.

Another thing -- are you aware of any more sophisticated tool than simple correlation coefficient, which would be able to do the things you are talking about, like catching "lagged" correlations?
5y
Juhani Huopainen Juhani Huopainen
As always, visual inspection of both the actual price data (in order to spot what effect the price level has in the relationship) and return data (in order to avoid seeing codependence when there is none). Cointegration is one statistical tool, as it looks beyond instantaneous correlation. One could construct a spread (long X units of oil, short Y units of whatever) and analyze the behavior of that spread - is it mean-reverting? Over what time frames? Also worth checking out is variance ratio-methodology.
5y
Juhani Huopainen Juhani Huopainen
The first line shows a hypothetical intraday chart of four trading days, with the closing prices marked with a red square. Notice how the price clearly varies, so there is plenty of variance in the market. The second line shows how the same closing prices, if viewed without the intraday data, shows no variance! If one would have calculated the variance from the intraday data, and then the daily data, a variance test would indicate that the variances (when adjusted for the different frequency) do not match, and that the first time series is in fact mean-reverting.
This is what was behind many of the early quantitative trading strategies. In fact, correlation does not help at all in predicting future prices, as correlation is by definition instantaneous. Correlation is only useful in constructing optimal portfolios. For price prediction, you need lead & lag!
Here’s one paper on variance ratios in action:
http://www.ljmu.ac.uk/AFE/AFE_docs/VR_AND_REGR.PDF
5y
DP DP
Let me take a look at the paper, and thanks again.

In a meantime, please take a look at the attached picture. It was quite fascinating finding for me, and I don't know if it has any applicable value. The lower indicator is ATR (Average True Range, which is measure of volatility of the tool). It's clearly seen on 15 min chart that there is amazing periodicity of volatility, which consists of major spike corresponding to appx 12:00 EST, and the smaller precursor to it around 4-5 AM EST. Major maximum corresponds to NY trading, and precursor, I believe, to Asian trading.

It's amazing for me how nice periodic volatility is, while analysed instrument is quite chaotic. I am wondering if one can make any use out of this observation.
5y
Juhani Huopainen Juhani Huopainen
Such very strong patterns point to obvious solutions: in this case it is the higher volatility at the time of the US market open (and for some time after the open). Another point is that when using averages like the ATR with period parameter set to 14, you must remember that the ATR reading presents an average of the past 13 data points and the current one - NOT only the current period.
Take a look of this chart I made you:
The WTI futures open outcry begins at 13:00 GMT and ends at 18:30 GMT (I hope I got those correctly).
5y
Juhani Huopainen Juhani Huopainen
In the middle panel I have the 14-period ATR, and in the bottom panel ATR with 1-period. Now you see that the volatility spike happens at the time of the market open, and volatility tends to remain high for some time, before cooling down. The 14-period ATR averages all these points and thus peaks only after the previous low-vol periods have dropped out of the 14-period range.

This U-shaped behavior of price volatility and trading volume is well-documented and exists in most markets. Most of the trading happens around the open and around the close. Higher amount of trading is associated with higher volatility. This creates some interesting paradoxes, like could financial market volatility be subdued if markets were open for a shorter time :)
5y
Juhani Huopainen Juhani Huopainen
There is no obvious way to make money by being able to predict volatility, as it is not a directly tradeable instrument. Option markets allow betting on volatility, but the market is good at discounting such well-known patterns beforehand, leaving nothing on the proverbial table.

One important lesson from this is that systematic or discretionary traders who trade intraday, should adjust their price projections and stop sizes based on the time of the day - larger stops and targets around open and close, and smaller ones during the middle of the session. As Bruce Lee said it, be like water ;)
5y
DP DP
Yes, you are right, and I had to notice that -- the N-point averaging shifts the indicator forward in time by N/2 point, i.e. 14-point ATR maximum is in fact located appx at 7.5*15min=1.8hr back in time, around 10 EST.

The only takeaway from ATR periodicity we can take is, as you say, stop distances for intraday trading.
5y
Juhani Huopainen Juhani Huopainen
I like these conversations - learning new stuff, formulating one's thoughts into words, debating...let's keep them coming! Some sort of GARCH-model that would take into account the known intraday patterns, in a sense adaptive and conditional Bollinger band-type indicator, sounds like an obvious addition to the too-long list of things to do.
5y
DP DP
Yes Junani, I like it too. Let's continue to do that!

Disclaimer

The Saxo Bank Group entities each provide execution-only service and access to Tradingfloor.com permitting a person to view and/or use content available on or via the website is not intended to and does not change or expand on this. Such access and use are at all times subject to (i) The Terms of Use; (ii) Full Disclaimer; (iii) The Risk Warning; (iv) the Rules of Engagement and (v) Notices applying to Tradingfloor.com and/or its content in addition (where relevant) to the terms governing the use of hyperlinks on the website of a member of the Saxo Bank Group by which access to Tradingfloor.com is gained. Such content is therefore provided as no more than information. In particular no advice is intended to be provided or to be relied on as provided nor endorsed by any Saxo Bank Group entity; nor is it to be construed as solicitation or an incentive provided to subscribe for or sell or purchase any financial instrument. All trading or investments you make must be pursuant to your own unprompted and informed self-directed decision. As such no Saxo Bank Group entity will have or be liable for any losses that you may sustain as a result of any investment decision made in reliance on information which is available on Tradingfloor.com or as a result of the use of the Tradingfloor.com. Orders given and trades effected are deemed intended to be given or effected for the account of the customer with the Saxo Bank Group entity operating in the jurisdiction in which the customer resides and/or with whom the customer opened and maintains his/her trading account. When trading through Tradingfloor.com your contracting Saxo Bank Group entity will be the counterparty to any trading entered into by you. Tradingfloor.com does not contain (and should not be construed as containing) financial, investment, tax or trading advice or advice of any sort offered, recommended or endorsed by Saxo Bank Group and should not be construed as a record of ourtrading prices, or as an offer, incentive or solicitation for the subscription, sale or purchase in any financial instrument. To the extent that any content is construed as investment research, you must note and accept that the content was not intended to and has not been prepared in accordance with legal requirements designed to promote the independence of investment research and as such, would be considered as a marketing communication under relevant laws. Please read our disclaimers:
- Notification on Non-Independent Invetment Research
- Full disclaimer

Check your inbox for a mail from us to fully activate your profile. No mail? Have us re-send your verification mail