I don't know how to do the last section. Please help
Look at the following dataset: (Data in StatPowers ) Sowce: Andrews, D.F. and Herzberg, A.M. (1985) Data: A Collection of Problems from Many Fields for the Student and Research Worker Springer Verlag. Description: 79 urine specimens were analyzed in an effort to determine if certain physical characteristics of the urine might be related to the formation of calcium oxalate crystals. Variables: r: Indicator of the presence of calcium oxalate crystals. gravity: The specific gravity of the urine. ph: The pll reading of the urine. osmo: The osmolarity of the urine. Osmolarity is proportional to the concentration of molecules in solution. cond: The conductivity of the urine. Conductivity is proportional to the concentration of charged ions in solution. urea: The urea concentration in millimoles per litre. calc: The calcium concentration in millimales per litre. (a) create a logistic model to predict the presence of calcium ocalate crystals based on the other six variables. Give the cocefficients of this full model: log odds(calcium-oxalate) - -355.338 355.944 "gravity + -0.496 -"ph + 0.017 Damo + -0.433 "cond + -0.032 urea + 0.784 "calc (b) Use backwards elimination to reduce the model to the model with the lowest AIC. Which is the first variable to eliminate? urea oph cond OSMO - calc - gravity (c) Repeat this process until you have a reduced model with the lowest AIC possible. Give the coefficients of the reduced model - put a zero in for any variable eliminated. log-odds(calcium-oxalate) - -500.011 + + 497.120 "gravity + 10-ph + 0 "osmo + -0.205 *cond + -0.018 *urea + 10.722 "calc (d) Suppose a urine sample has gravity-1.018, ph-5.29, osmo 698, cond-28, urea 252 and calc H. Use the first (full) model to predict the probability this patient will have calcium oxalate crystals. First, what is the log odds ratio? (e) Now convert the log-odds ratio to a probability. Use the formula pre 'Alte') where Y is the predicted log odds ratio. (enter a decimal between 0 and 1) (f) Now find the probability using the reduced model (enter a decimal between 0 and 1)