Lift berekenen & significantie testen

Marketingcampagnes analyseren met pandas

Jill Rosok

Data Scientist

Treatmentprestaties vs. controle

Lift berekenen:

$$ \frac{\text{Conversie treatment - conversie controle}}{\text{Conversie controle}} $$

Lift berekenen

# Calcuate the mean of a and b 
a_mean = np.mean(control)
b_mean = np.mean(personalization)

# Calculate the lift using a_mean and b_mean
lift = (b_mean-a_mean)/a_mean

print("lift:", str(round(lift*100, 2)) + '%')

lift: 194.23%

T-verdeling

¹ Identification of Timed Behavior Models for Diagnosis in Production Systems. Scientific Figure on ResearchGate.

P-waarden

T-statistiek van 1,96 is meestal significant op 95%-niveau
Afhankelijk van de context kies je soms een lager of hoger significantieniveau.

T-toets in Python

from scipy.stats import ttest_ind

t = ttest_ind(control, personalized)

print(t)

TtestResult(statistic=-2.7343299447505074, 
            pvalue=0.006451487844694175, 
            df=552.0)

Laten we oefenen!

Marketingcampagnes analyseren met pandas