def ModeratedWelchTest(gs,d0,s0_squared):
df = gs.DF()
### use the s0_squared for the pairwise comparison calculated in the getModeratedStandardDeviation
svar1 = (d0*s0_squared+df*gs.Svar1())/(d0+df)
svar2 = (d0*s0_squared+df*gs.Svar2())/(d0+df)
#svar = sg ### Use this to test and see if this gives the same result as a non-moderated t-test
if svar1 != 0 and svar2 != 0:
t = (gs.Avg1()-gs.Avg2())/math.sqrt(svar1+svar2)
prob = salstat_stats.betai(0.5*df,0.5,float(df)/(df+t*t))
else: prob = 1
#gs.SetAdjP(prob)
gs.setPval(prob)
def ModeratedWelchTest(gs,d0,s0_squared):
df = gs.DF()
### use the s0_squared for the pairwise comparison calculated in the getModeratedStandardDeviation
svar1 = (d0*s0_squared+df*gs.Svar1())/(d0+df)
svar2 = (d0*s0_squared+df*gs.Svar2())/(d0+df)
#svar = sg ### Use this to test and see if this gives the same result as a non-moderated t-test
if svar1 != 0 and svar2 != 0:
t = (gs.Avg1()-gs.Avg2())/math.sqrt(svar1+svar2)
prob = salstat_stats.betai(0.5*df,0.5,float(df)/(df+t*t))
else: prob = 1
#gs.SetAdjP(prob)
gs.setPval(prob)
def incompleteBeta(t,df):
p = salstat_stats.betai(0.5*df,0.5,df/(df+t*t))
return p
def incompleteBeta(t,df):
p = salstat_stats.betai(0.5*df,0.5,df/(df+t*t))
return p
