コモディティ・為替・先物データで景気の風向きを読む：原油・金・ドル円・VIXを社内KPIに重ねる

!pip install -q yfinanceimport yfinance as yfimport matplotlib.pyplot as pltimport japanize_matplotlib  # noqa: F401
# WTI 原油（CL=F は USD/バレル）と USD/JPYoil = yf.Ticker("CL=F").history(period="5y")["Close"]usd = yf.Ticker("JPY=X").history(period="5y")["Close"]
# 原油を「円建て」に変換（≒ 国内の輸入原油価格の代理指標）oil_jpy = (oil * usd).dropna()
fig, ax = plt.subplots(figsize=(12, 5))ax.plot(oil_jpy.index, oil_jpy.values, color="#d62728", linewidth=1.2,        label="WTI 原油（円建て換算）")ax.set_title("WTI 原油の円建て価格（円安が効くと価格上昇が増幅）")ax.set_ylabel("円/バレル"); ax.legend(); ax.grid(alpha=0.3)plt.tight_layout(); plt.show()

copper = yf.Ticker("HG=F").history(period="5y")["Close"]n225   = yf.Ticker("^N225").history(period="5y")["Close"]
# 開始日=100 で正規化copper_norm = copper / copper.iloc[0] * 100n225_norm   = n225 / n225.iloc[0] * 100
fig, ax = plt.subplots(figsize=(12, 5))ax.plot(copper_norm.index, copper_norm.values, label="銅 HG=F", color="#b87333", linewidth=1.5)ax.plot(n225_norm.index, n225_norm.values, label="日経225", color="#1f77b4", linewidth=1.5)ax.set_title("銅 vs 日経225（5年・開始日=100で正規化）")ax.legend(); ax.grid(alpha=0.3); plt.tight_layout(); plt.show()

vix = yf.Ticker("^VIX").history(period="5y")["Close"]
fig, ax = plt.subplots(figsize=(12, 5))ax.plot(vix.index, vix.values, color="#1f77b4", linewidth=1)ax.axhspan(0,  20, alpha=0.1, color="green", label="平穏（〜20）")ax.axhspan(20, 30, alpha=0.1, color="yellow", label="警戒（20〜30）")ax.axhspan(30, 100, alpha=0.1, color="red", label="動揺（30〜）")ax.set_title("VIX（恐怖指数）の推移と危険水準")ax.set_ylabel("VIX"); ax.legend(); ax.grid(alpha=0.3)plt.tight_layout(); plt.show()

import pandas as pdimport seaborn as sns
# 月次に揃えるdef monthly(ticker):    return yf.Ticker(ticker).history(period="5y")["Close"].resample("MS").last()
df = pd.DataFrame({    "TOPIX":   monthly("^TPX"),    "原油":    monthly("CL=F"),    "銅":      monthly("HG=F"),    "USDJPY":  monthly("JPY=X"),    "VIX":     monthly("^VIX"),    "米10年金利": monthly("^TNX"),    "自社売上": ...  # 社内 DWH から取得})df_pct = df.pct_change().dropna()
corr = df_pct.corr()plt.figure(figsize=(8, 6))sns.heatmap(corr, annot=True, fmt=".2f", cmap="RdBu_r", center=0,            vmin=-1, vmax=1, square=True)plt.title("マクロ指標 × 自社売上の月次変化率の相関")plt.tight_layout(); plt.show()