Multiple regression using OLS from stats model

Question

I have looked up similar questions to mine, I cannot find an answer.

My aim: I have survival data. I want the residuals for the survival data, after accounting for age and weight.

Method:

import statsmodels
from statsmodels import ols
species = ["sample1","sample2","sample3","sample4","sample5"]
Survival = [0.1,0.2,0.3,0.4,0.5]
Age = [1,2,3,4,5]
Weight = [1,3,5.5,7,10]
mymodel = ols.ols(Mortality,[Weight,Age],"Mortality",["Weight","Age"])
print mymodel

Output: My ideal output is a table with two columns, one column being the species, the other column being the mortality residual after I have accounted for age and weight.

Questions: 1.No matter what I do, I cannot find the ols method. I have statsmodel installed. When I open a python console, and do dir(statsmodels), there are the options that I get:

['CacheWriteWarning', 'ConvergenceWarning', 'InvalidTestWarning', 'IterationLimitWarning', 'NoseWrapper', 'Tester', '__builtins__', '__doc__', '__docformat__', '__file__', '__name__', '__package__', '__path__', '__version__', 'compat', 'datasets', 'distributions', 'errstate', 'print_function', 'simplefilter', 'test', 'tools', 'version']

Where is the ols method?

2. Hopefully once I can actually find the method, I will be able to run the code and ask where I can find the residuals that I'm looking for.

Thanks

Answer 1

Thank you, that worked.

Here is the code that I used:

import numpy as np
import statsmodels.api as sm
import statsmodels.formula.api as smf
import sys

dat = np.loadtxt(sys.argv[1],dtype={"names":("Species","Weight","Mortality","Age"),"formats":("S20","f4","f4","f4")})
mymodel = smf.ols("Mortality~Weight+Age",data=dat).fit()
Residues = list(mymodel.resid_pearson)
for each_species,each_residue in zip(list(dat["Species"]),Residues):
    print each_species + "\t" + str(each_residue)

So I think/hope what is happening here is that I'm reading in the table, fitting a multiple regression to the data, and printing the species name and the residual for each species (the residual being the regression of mortality against weight and age). Thank you.