Calibration and verification of risk algorithms using logistic regression

The use of logistic regression is proposed as a method of verifying and calibrating disease risk algorithms. The logistic regression model calculates the log of the odds of a binary outcome as a function of a linear combination of predictors. The resulting model assumes a multiplicative (relative) relationship between the different risk factors. Computer programs for performing logistic regression produce both estimates and standard errors, thus permitting the evaluation of the importance of different predictive variables. The use of receiver operating characteristic (ROC) curves is also proposed as a means of comparing different algorithms. An example is presented using data on Sclerotinia stem rot in oil seed rape, caused bySclerotinia sclerotiorum.

[1]  J. Yuen,et al.  Forecasting Sclerotinia stem rot in spring sown oilseed rape , 1998 .

[2]  A. Agresti.  Categorical data analysis , 1993 .

[3]  R. A. Krause,et al.  Blitecast: a computerized forecast of potato late blight [Phytophthora infestans, fungus diseases]. , 1975 .

[4]  M. Jeger,et al.  The influence of soil temperature, soil moisture, soil texture, and inoculum density on the incidence of sorghum downy mildew. , 1987 .

[5]  T. Gillespie,et al.  A predictive scheme for timing fungicide applications to control alternaria leaf blight in carrots , 1979 .

[6]  Ross Darnell,et al.  A model for estimating infection levels of anthracnose disease of mango , 1984 .

[7]  Dennis A. Johnson,et al.  Weather and downy mildew epidemics of hop in Washington state. , 1994 .

[8]  D. McCracken,et al.  Factors affecting the size of leatherjacket (Diptera:Tipulidae) populations in pastures in the west of Scotland , 1995 .

[9]  R. A. Krause.  Blitecast; a computerized forecast of potato late blight. implementation and evaluation , 1976 .

[10]  John C. Dodd,et al.  The effect of climatic factors on Colletotrichum gloeosporioides, causal agent of mango anthracnose, in the Philippines. , 1991 .

[11]  D. Hosmer,et al.  Applied Logistic Regression , 1991 .

[12]  P. McCullagh,et al.  Generalized Linear Models , 1992 .

[13]  W. Schuh.  Effect of pod development stage, temperature, and pod wetness duration on the incidence of purple seed stain of soybeans. , 1992 .

[14]  C. Metz.  Basic principles of ROC analysis. , 1978, Seminars in nuclear medicine.

[15]  D. Hosmer,et al.  Applied Logistic Regression , 1991 .