Using the patented Coles2/JXBR short-term property price projection methodology (CJSTPPPM) based on a lag between consumer sentiment and the rolling year on year % price change, I did the following.
I got Canadian consumer sentiment data from the OECD - conferenceboard.ca/ and I am not that interested). The only series available for a long duration is quarterly, from Q2-1980 to Q2-2012. There is a monthly series but it is only available from Jan-2002 to Jul-2012. A simple linear interpolation of the quarterly values agreed closely with the month values - R-squared of 0.9579 - so I went with the converted quarterly values.
This applies to the property index series “Whole Country Single-Family Houses New National Statistical Office Per Square Meter Index, 2007 = 100” for all of Canada and so is very generalised and does not take into account local variations.
The best correlation between the consumer sentiment and the year on year % price change is obtained at a lag of 6 months - R-squared of 0.7657. Not as high as I would have liked but it is the best I could get.
Notwithstanding previous comments made elsewhere about time series heteroskedasticity and unexpected external interventions, the outcome from this is:
Consumer sentiment dipped slightly in the last 6 months so based on this magical technique, Canadian property prices should either fall slightly but not significantly or remain at their current levels with localised variations.
Thank you folks. I am available for birthday parties, bar mitzvahs, bat mitzvahs and other special occasions.