Leverage dan conviction

Analisis Market Basket dengan Python

Isaiah Hull

Visiting Associate Professor of Finance, BI Norwegian Business School

Membangun dari metrik sederhana

$$Support(X) = \frac{Frequency(X)}{N}$$

$$Support(X \rightarrow Y) = \frac{Frequency(X \& Y)}{N}$$

$$Confidence(X \rightarrow Y) = \frac{Support(X \rightarrow Y)} {Support(X)}$$

$$Lift(X \rightarrow Y) = \frac{Support(X \rightarrow Y)}{Support(X) Support(Y)}$$

Metrik leverage

Leverage juga dibangun dari support.

$$Leverage(X \rightarrow Y) = $$ $$Support(X \& Y) - Support(X) Support(Y)$$

Leverage mirip lift, tetapi lebih mudah ditafsirkan.
Nilai leverage berkisar -1 hingga +1.
- Lift berkisar dari 0 hingga tak hingga.

Menghitung leverage

# Hitung support untuk Twilight dan Harry Potter
supportTP = np.logical_and(books['Twilight'], books['Potter']).mean()

# Hitung support untuk Twilight
supportT = books['Twilight'].mean()

# Hitung support untuk Harry Potter
supportP = books['Potter'].mean()

# Hitung dan cetak leverage
leverage = supportTP - supportP * supportT
print(leverage)

0.018

Metrik conviction

Conviction juga dibangun dari support.
Lebih rumit dan kurang intuitif dibanding leverage.

$$Conviction(X \rightarrow Y) = $$ $$\frac{Support(X) Support(\bar{Y})} {Support(X \& \bar{{Y}})}$$

Menafsirkan conviction

Gambar mini Twilight.

Gambar mini Harry Potter and the Sorcerer's Stone.

Gambar mini Twilight.

Gambar mini The Hunger Games.

¹ Gambar diambil dari goodreads.com.

Menghitung conviction

# Hitung support untuk Twilight dan Harry Potter serta Twilight
supportTP = np.logical_and(books['Twilight'], books['Potter']).mean()
supportT = books['Twilight'].mean()

# Hitung support untuk BUKAN Harry Potter
supportnP = 1.0 - books['Potter'].mean()

# Hitung support untuk Twilight dan BUKAN Harry Potter
supportTnP = supportT - supportPT

# Hitung conviction
conviction = supportT*supportnP / supportTnP
print(conviction)

1.16

Ayo berlatih!

Analisis Market Basket dengan Python