Canada Markets

Statistics Canada Reveals New Forecasting Methodology

Cliff Jamieson
By  Cliff Jamieson , Canadian Grains Analyst
Connect with Cliff:
Statistics Canada's new methodology for estimating Canadian crop production saw production prospects increase for many of Canada's crops from earlier July estimates. This chart compares the new model-based estimates (blue bars) with the July estimates (red bars) along with 2014 production (green bars) and the five-year average (purple bars) for selected crops. (DTN graphic by Nick Scalise)

Statistics Canada revealed its first-ever model-based principal field crop estimates Thursday. Statistics Canada combined its Crop Condition Assessment Program (based on satellite data), along with its regular Field Crop Reporting Series based on producer surveys, as well as agroclimatic data.

This year's Sept. 17 report falls between the survey-based July estimates and with the September Farm Survey estimates, which will be released in the production of principal field crops report on Oct. 2. Both rely on conventional Statistics Canada methodology.

I was contacted by a Saskatchewan producer on the same day as this release. He was concerned about the challenges to estimate production. These included: some countries have incentives to overstate production to obtain subsidies; global stockpiles are at risk because of quality deterioration; and domestic producers intentionally feed inaccurate information on surveys. Some people also have a lack of confidence in the government's reports.

While this particular report may be the first of its kind, Statistics Canada has used satellite data to estimate yields in the past. In 2012, satellite data was used to produce yield estimates which were higher than final official Statistics Canada yields. The Crop Condition Assessment Program (CCAP) released yield estimates that were .7% higher than the final official Statistics Canada data for spring wheat, 21.7% higher for canola, .8% higher for durum and 14.4% higher for barley. Across these four crops reported, average yields estimated by the CCAP were 9.4% higher than final official estimates.

P[L1] D[0x0] M[300x250] OOP[F] ADUNIT[] T[]

In 2013, CCAP estimates were viewed as too low, in fact, The Western Producer reported them to be "way off the mark" according to trade responses. Prairie spring wheat yields reported by CCAP were 17.4% lower than the final official estimate, while 15% lower for canola, 18.4% lower for durum and 12.6% lower for barley. Across the four commodities, average yields estimated by the CCAP were 15.9% lower than the final reported estimate.

2014 data once again saw CCAP estimates come in below the final official estimates based on producer surveys. CCAP's spring wheat yield was 11.4% below the final estimate, the canola yield was 1.7% below the final estimate, the durum yield was 10.5% below the final estimate and the barley estimate was 12.9% below the final official estimate. Across the four commodities, the average CCAP yield estimate was 9.1% below the final official estimates.

The attached chart indicates many Canadian crops improving since the July estimates were released in August, as seen in the move from the red bar to the blue bar. This is widely expected because of more favorable weather seen in late summer. Of the crops shown, only oats and corn are expected to see production increase from last year, while durum, oats and corn crops are expected to reach production levels above their five-year averages.

On Oct. 2, survey-based results will be reported and this week's data release can be compared to data obtained from conventional methods.


DTN 360 Poll

This week's poll asks if you have booked fertilizer for 2016? You can weigh in with your thoughts on DTN's 360 poll which is found on your DTN homepage.

Cliff Jamieson can be reached at cliff.jamieson@dtn.com

Follow Cliff Jamieson on Twitter @CliffJamieson

(ES)

P[] D[728x170] M[320x75] OOP[F] ADUNIT[] T[]
P[L2] D[728x90] M[320x50] OOP[F] ADUNIT[] T[]

Comments

To comment, please Log In or Join our Community .