

In [2]:

    
%matplotlib inline
import matplotlib.pyplot as plt



In [3]:

    
from vespa.stars import MultipleStarPopulation



In [4]:

    
from scipy.stats import powerlaw
mmin=0.2; mmax=0.5
ms = powerlaw(3).rvs(int(1e5)) * (mmax-mmin) + mmin #rough approximation to the mass distribution of Kepler M's
plt.hist(ms);



In [27]:

    
pop = MultipleStarPopulation(mA=ms, qmin=0.2, max_distance=500, minmass=0.15)



In [28]:

    
query = 'r_mag < 17'
frac = pop.binary_fraction(query) + pop.triple_fraction(query)



In [29]:

    
frac/.52









    Out[29]:





1.2463947259991759



In [31]:

    
plt.hist(pop.stars['distance'], histtype='step', lw=3)
plt.hist(pop.stars.query(query)['distance'].values, histtype='step', lw=3);



In [35]:

    
sub = pop.stars.query(query)
(sub['radius_A'] + sub['radius_B']).mean() / (pop.stars['radius_A'] + pop.stars['radius_B']).mean()









    Out[35]:





1.079022257298867

OK, so for multiple systems, average eclipse probability is about 8% higher than for volume-limited sample. And binary fraction is about 20-25% higher. What does this mean for an appropriate correction factor? Haven't thought this through carefully yet.



In [ ]: