Fundamentally Speaking

Corn Crop Ratings vs. Yield Correlations

Joel Karlin
By  Joel Karlin , DTN Contributing Analyst

There has always been interest in the USDA's weekly condition report for the various crops especially with the absolute trend and how they change during the growing season.

We, along with others, use these crop ratings as one of the variables in our crop yield model along with certain weather data and crop progress indications.

This graphic shows the correlation coefficient for corn crop ratings at the end of July, end of August and end of September vs. the percent that final corn yields deviated from the 1986-2014 trend for each of the top 18 corn producing states and the U.S.

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

The only exceptions are for North Dakota and Tennessee since these two states did not become part of the top 18 state group till the year 2000 so we use crop ratings and yields for the period 2000-2014.

We use our usual ratings system where we weight the crop based on the percent in each category and assign that category a factor of 2 for very poor, 4 for poor, 6 for fair, 8 for good, and 10 for excellent and then sum the results.

For whatever reason, the top states in the Plains region such as Colorado, Kansas, North Dakota and Texas have lower correlation coefficients than the Midwest or Delta states.

Most of the states have correlation coefficients above the 0.80 level and some are even over the 0.90 level indicating a strong level of correlation.

We had also thought that the correlation coefficient would increase as the crop year went on with the coefficient higher for the end of August ratings as opposed to the end of July ratings and the end of September ratings having an even higher correlation coefficient vs. deviation or final yields than the end of August ratings.

This occurred in 11 of the 19 cases with the correlation coefficient actually declining over time in Texas.

(KA)

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 .