Posts Tagged ‘quantitative analysis’

XBB and correlation between Canadian Stocks & Canadian Bonds

March 13, 2011 Leave a comment

XBB – Conflicting technical signals

  • XBB broke out of recent downtrend line but… not above the previously broken long-term uptrend line
  • RSI & MACD are rising but price is declining
  • Also death cross signal – 50 day SMA crossed 200 day SMA from above

Fundamentally, Canadian interest rates are bound to rise in the near future which is not good for bonds.

Conclusion…keep a close watch for the next couple weeks

What is the correlation between returns on Canadian stocks and Canadian bonds?

I’m going to use the bond ETF, XBB to represent the DEX Universe Bond Index and S&P TSX Composite to represent stocks.

Correlation of Daily Returns is:

  • fairly volatile for smaller samples i.e. less than 3 months. (blue & green lines)
  • low and positive from 2001 to March 2010
  • -0.2 since March 2010 i.e. when one goes up, the other goes down.

Lets look at the correlation of monthly returns… partly because the daily volatility has been much higher since the financial crisis and that might bias the results. Again, it is a similar story, the smaller the sample, the greater tha

Correlation of monthly returns is similar to daily returns except:

  • 12 month rolling correlation deviates between 0.5 and -0.5 which is on the high side.
  • 3 yr correlation has been positive and low, around 0.1 since Jan 2009.


Rising Interest Rates – thanks to QE?

December 16, 2010 Leave a comment

If you haven’t already heard so…interest rates have reversed their downward trend to dramatically move up since the November Federal Reserve meeting, Quantitative Easing (QE2) announcement… QE2 is supposed keep US interest rates low… would you have thought that it would affect rates elsewhere?

Year-to-Date chart of Canadian interest rates…

As marked on the chart, interest rates in Canada have risen significantly over the last 6 weeks… especially the medium to long term rates in the 2-10 year terms. The 5-year rate is at 2.56%, same as in mid-July.

Here is the term chart or the Canadian yield curve… see the parallel shift in yield across all terms!

This means, borrowing costs will increase in proportion… yes, including mortgage rates, particularly fixed rates.

The 5-year fixed rates are as low as they have ever been… but they will be rising shortly; see this

The Bank of Canada hasn’t indicated any shift in monetary policy since the last rate hike in Sep-2010… so why are rates in Canada rising? Possible reasons:

· Higher inflation expectation

· Better than expected economic growth

· Bond markets are overbought

· Rising risk of default (!)

Stay tuned… I will explore each of these possibilities in the coming days…

World Stock Markets near 52 Week Highs

October 26, 2010 Leave a comment

Close to overbought territory… Japanese equity markets clobbered due to 15-year high Yen.

World Equity Markets – Performance from 52 week high/low

September 21, 2010 2 comments

Equity markets in India and Canada are just shy of their 52-week highs! Chinese stocks are struggling to reach new highs and could very well be on their way down

Actually, the SENSEX has reached new 52-week highs in recent days and is up 25% from its May 2010 lows.

TSX is up 5% in 7 days – will the rally continue or fizzle out?

September 3, 2010 2 comments

The TSX Composite index is up 4.79% in the last 7 trading sessions on increasing volume… quite the gain you might say. What is the return following a huge parabolic gain?  

First, here is the 3-year chart showing the 7-day return…  



Clearly there are few days where the 7-day return is greater than 5%… I suspect that is due to  higher than average volatility… So I went back and looked at 10 years of data and is here the distribution of 7-day returns  


Return Period – 7 days  

MIN – -22.6%  

MAX – 15.8%  

Threshold – 5%  

Observations – 2686  

Observations > Threshold – 99  

Just under 4% or 100 days in the last 10 years was the 7-day return greater than 5%… What is the 3-day return after these 99 observations? (I choose 3-days simply because it takes a lot less time to go down than to go up…yes, I’m biased to the downside)  

And here is a distribution of the 3-day (ex-post) returns after the 7-day 5% rally… the positive vs negative periods are fairly distributed (52% negative vs 48% positive) but with a fatter right tail i.e. more periods with higher positive returns or in other words… the chances of higher gains are higher than higher losses.  



   Note: As of 10am today (03Sep2010) the TSX is up about 0.35%

Short term uptrend in Gold broken, Correlation to USD & more

July 31, 2010 4 comments

The 1-year uptrend channel (blue lines) is violated… The recent new (nominal) high did not make a new high in RSI or MACD…(negative divergence for the technically inclined & the dotted red lines on the chart) means the recent high was made on declining momentum – not a good sign.

I have drawn out the various support levels from here down and my short term (< 1 year) bias is to the downside especially for August… Demand is slow during the summer months because most jewelery manufacturers are shut down and don’t reopen till September…

Gold chart with technical analysis

US Dollar & Gold Correlation

In theory, a commodity priced in USD dollars will move inversely to the value of USD e.g.- if Gold goes up in USD, then the value of USD in other currencies goes down and vice versa. Among all commodities, Gold particularly exhibits a strong negative correlation in a “normal” volatility year… during highly volatile or uncertain economic periods, this correlation deviates significantly from the norm…

Lately Gold & USD have been moving in lock-step… i.e. positively correlated as illustrated in this nice chart from Bespoke Investment Group

Chart showing correlation of Gold & USD

Source: Bespoke Investment Group

Eventually the correlation has to revert to its mean… which would mean that USD & Gold move in opposite directions. Couple weeks ago I posted a chart of the US Dollar Index positing that USD will reverse the downtrend and rally… hasn’t happened yet but I still think the USD is at a key support level and will rally from here and expect Gold to continue the downtrend…

Also, Gold has risen 21% from 1044 in Feb 2010 to its intraday high of 1265 in June 2010 without a 10% correction… of course it can go higher without correcting but the subsequent correction will be just as severe…

Lastly… Inflation

Historically, Gold is also meant to be a hedge against inflation… if you have been reading the news lately, you know well that prices are declining not rising so I believe the argument of buying Gold right now as a hedge against inflation is bogus!


Update: Gold during Deflation via ZeroHedge

According to Casey Reseach

There’s lots of data about what gold does during periods of high inflation, but less so with deflation, partly because we don’t see a true deflation all that often. But of course we’ve got the biggie we can look at, and the seriousness of the Great Depression can give us a big clue as to how gold stocks behave in a true deflationary environment

This chart from the above post is quite interesting:


From 1929 until January 1933, the stock of Homestake Mining, the largest gold producer in the U.S., rose 474%. Dome Mines, the largest Canadian producer, advanced 558%. In spite of the gold price being fixed at the time, gold stocks rose dramatically.

At the same time, the DJIA lost 73% of its value

The bottom line is that the two largest gold producers – during a time of soup lines and falling standards of living – handed investors five and six times their money in four years.

What about gold itself? On April 5, 1933, President Roosevelt issued an executive order forcing delivery (i.e., confiscation) of gold owned by private citizens to the government in exchange for compensation at the fixed price of $20.67/oz (you can read the original order here). And less than nine months later, he raised the gold price to $35, effectively diluting every dollar 41% overnight and swindling everyone who had turned in his gold.We don’t know exactly what an untethered gold price would have done during the depression, but given its distinction in history as a store of value, we believe it would retain its purchasing power in a deflationary setting regardless of its nominal price. In other words, while the price of gold might not rise, or could even fall, your best protection is still gold.


 This is the only quasi-convincing article I have seen on owning Gold in deflation. A lot of well-respected economists (David Rosenberg, Roubini, etc) and financial bloggers (Mish) are urging investors to own gold during deflation but without an objective reasoning… Perhaps that is as noted above due to the lack of true deflation in a fiat monetary system.

Death Cross on the S&P TSX Composite?

July 16, 2010 3 comments

The TSX is awefully close to a Death cross signal (50 day simple moving average crossing the 200 day moving average from above i.e. 50 day average is less than 200 day average)…

I think we are bound to get a death cross within the next week… the TSX is about 300 points away to signal the death cross as of yesterday’s close.

Technical Analysis of S&P TSX Composite Index with Death Cross


How good is the Death Cross signal? To answer this I looked at all the historical data I could get and analyzed the returns based on Short Selling at close on the day of Death Cross and covering and going Long at the next Golden cross (50 day SMA crosses 200 day SMA from below)

I understand 10 years is a small sample but it is better than nothing 🙂

If you only traded on the Death Cross & Golden Cross from the first signal in Nov 2000 you would be way ahead of the market for the analyzed period analyzed… hell, you would be better off just going long and sitting out until a Golden cross is recognized!

On the other hand, the Death Cross is only good half the time where as the golden cross is only good 4 out of 10 times BUT provides a better return than the death cross…

24-Nov-2000 15-Jul-2010
Total Death Cross Golden Cross
Short Trade Long Trade
Total Trades 11 6 5
Winning Trades 5 3 2
% Winning Trades 45% 50% 40%
Largest Winning Trade 56.8% 20.5% 56.8%
Largest Losing Trade -7.7% -7.1% -7.7%
Cumulative Compounded Return 125.0% 25.8% 78.9%
$1000 = 2,249.82 1,257.75 1,788.77
Buy & Hold Return 30%
$1000 = 1,301.11

(If you know where I can get historical data beyond 2000, please let me know)

For more information on Death Cross & Golden cross check out: