Space73 / app.py
QuantumLearner's picture
Create app.py
c1fa6c7 verified
import streamlit as st
import numpy as np
import pandas as pd
import plotly.graph_objects as go
from datetime import datetime, timedelta
import yfinance as yf
# Set wide page layout and page title
st.set_page_config(layout="wide", page_title="Self-Tuning SuperTrend and K-Means")
# App title and purpose explanation
st.title("Self-Tuning SuperTrend and K-Means")
st.write("This tool builds a self-tuning SuperTrend indicator using k-means clustering to adapt stop levels and generate buy/sell signals. It uses daily price data to compute volatility, score multiple configurations, and select the best one in real time.")
# Methodology expander (closed by default)
with st.expander("Methodology", expanded=False):
#st.write("The tool self-tunes the SuperTrend indicator by testing multiple configurations and picking the best one.")
#st.write("**Step 1: Data Preparation**")
#st.write("Daily price data is downloaded. Key columns (High, Low, Close) are retained and prepped.")
st.write("**Volatility Measurement**")
st.write("An Average True Range (ATR) is computed using an Exponential Moving Average (EMA).")
st.latex(r'\text{ATR} = \text{EMA}\left(\max\left\{High-Low,\ |High-\text{PrevClose}|,\ |Low-\text{PrevClose}|\right\}\right)')
st.write("**Generating SuperTrend Variants**")
st.write("Multiple SuperTrend signals are calculated. They use:")
st.latex(r'\text{Upper Band} = hl2 + ATR \times \text{factor}')
st.latex(r'\text{Lower Band} = hl2 - ATR \times \text{factor}')
st.write("**Performance Scoring & Clustering**")
st.write("Each variant is scored based on price movement. K-means (k=3) clusters these scores into Best, Average, and Worst groups.")
st.write("**Final Signal Generation**")
st.write("The indicator is recomputed using the average factor from the selected cluster. This gives a self-calibrated trading signal.")
st.write("This process minimizes manual tuning and adapts with recent price action.")
st.write("For more details, see the this article [here](https://entreprenerdly.com/trading-signals-with-adaptive-supertrend-and-k-means/).")
# Sidebar inputs explanation
st.write("#### Adjustable Inputs & Implications")
st.write("""
- **Ticker:** The stock symbol to analyze. Changing this lets you switch assets.
- **Start Date & End Date:** Define the analysis window. End Date defaults to today plus one day.
- **ATR Length:** Sets the period for ATR. Lower values react faster; higher values smooth out noise.
- **Minimum/Maximum Multipliers & Step:** Define the range for SuperTrend sensitivity. Smaller steps improve resolution but increase compute time.
- **Performance Alpha:** Determines the smoothness of the performance score. Lower makes the metric more reactive; higher favors stability.
- **K-Means Options (From Cluster & Max Iterations):** Choose which cluster (Best, Average, Worst) to use and limit iterations for clustering. This controls how variants are grouped.
""")
# Sidebar inputs
with st.sidebar:
st.header("Input Parameters")
# Data inputs expander
with st.expander("Data Inputs", expanded=True):
ticker = st.text_input(
"Ticker",
value="ASML",
help="Enter the stock symbol to analyze. Example: AAPL, MSFT, NVDA. This determines which asset's data will be used."
)
start_date = st.date_input(
"Start Date",
value=datetime(2022, 1, 1),
help="Start of the historical data window. Affects the amount of price history used to compute signals."
)
default_end_date = datetime.today() + timedelta(days=1)
end_date = st.date_input(
"End Date",
value=default_end_date,
help="End of the data window. Automatically set to today + 1 to include the most recent bar."
)
# Methodology parameters expander
with st.expander("Methodology Parameters", expanded=True):
atr_length = st.number_input(
"ATR Length",
min_value=1,
value=7,
step=1,
help="ATR period controls how volatility is measured. Lower values make the stop more sensitive to short-term moves. Higher values smooth noise but may react slower."
)
min_mult = st.number_input(
"Minimum Multiplier",
value=1.0,
step=0.1,
help="Defines the tightest SuperTrend stop. Lower values mean tighter stops, which react quickly but may whipsaw in choppy conditions."
)
max_mult = st.number_input(
"Maximum Multiplier",
value=5.0,
step=0.1,
help="Defines the widest SuperTrend stop. Higher values give more breathing room but may delay trend changes."
)
step_mult = st.number_input(
"Step",
value=0.5,
step=0.1,
help="Step size between multipliers. Smaller values give finer resolution but increase compute time."
)
perf_alpha = st.number_input(
"Performance Alpha",
min_value=1,
value=8,
step=1,
help="Controls how quickly the performance metric responds to new price behavior. Lower = more reactive, higher = more stable but slower to adapt."
)
from_cluster = st.selectbox(
"From Cluster",
options=["Best", "Average", "Worst"],
help="Selects which k-means cluster to use for final signal generation. 'Best' is typical for trend-following. 'Average' or 'Worst' can simulate conservative or contrarian behavior."
)
max_iter = st.number_input(
"Max Iterations",
min_value=1,
value=1000,
step=1,
help="Upper limit on how long k-means clustering can run. Higher values allow more precise convergence but slow down the run time."
)
# Action button
run_analysis = st.button("Run Analysis")
if run_analysis:
# Validate date input
if start_date >= end_date:
st.error("Start Date must be before End Date.")
else:
with st.spinner("Running analysis..."):
try:
# Convert dates to string format
start_date_str = start_date.strftime("%Y-%m-%d")
end_date_str = end_date.strftime("%Y-%m-%d")
# 1) Download data
df = yf.download(ticker, start=start_date_str, end=end_date_str, interval="1d", auto_adjust=False)
if df.empty:
st.error("No data returned from the data provider.")
st.stop()
if isinstance(df.columns, pd.MultiIndex):
df.columns = df.columns.get_level_values(0)
df.rename(columns={"Open": "Open", "High": "High", "Low": "Low", "Close": "Close", "Volume": "Volume"}, inplace=True)
df.dropna(subset=["High", "Low", "Close"], inplace=True)
df["hl2"] = (df["High"] + df["Low"]) / 2.0
# 2) Compute ATR
df["prev_close"] = df["Close"].shift(1)
df["tr1"] = df["High"] - df["Low"]
df["tr2"] = (df["High"] - df["prev_close"]).abs()
df["tr3"] = (df["Low"] - df["prev_close"]).abs()
df["tr"] = df[["tr1", "tr2", "tr3"]].max(axis=1)
df["atr"] = df["tr"].ewm(alpha=2/(atr_length+1), adjust=False).mean()
df.dropna(inplace=True)
df.reset_index(drop=False, inplace=True)
n = len(df)
# Helper function: sign
def sign(x):
return np.where(x > 0, 1, np.where(x < 0, -1, 0))
# 3) Compute supertrend for each factor
def compute_supertrend(df, factor, perf_alpha):
arr_close = df["Close"].values
arr_hl2 = df["hl2"].values
arr_atr = df["atr"].values
trend = np.zeros(n, dtype=int)
upper = np.zeros(n, dtype=float)
lower = np.zeros(n, dtype=float)
output = np.zeros(n, dtype=float)
perf = np.zeros(n, dtype=float)
trend[0] = 1 if arr_close[0] > arr_hl2[0] else 0
upper[0] = arr_hl2[0]
lower[0] = arr_hl2[0]
output[0] = arr_hl2[0]
perf[0] = 0.0
for i in range(1, n):
up = arr_hl2[i] + arr_atr[i] * factor
dn = arr_hl2[i] - arr_atr[i] * factor
if arr_close[i] > upper[i-1]:
trend[i] = 1
elif arr_close[i] < lower[i-1]:
trend[i] = 0
else:
trend[i] = trend[i-1]
if arr_close[i-1] < upper[i-1]:
upper[i] = min(up, upper[i-1])
else:
upper[i] = up
if arr_close[i-1] > lower[i-1]:
lower[i] = max(dn, lower[i-1])
else:
lower[i] = dn
diff_sign = sign(arr_close[i-1] - output[i-1])
perf[i] = perf[i-1] + 2/(perf_alpha+1)*((arr_close[i] - arr_close[i-1]) * diff_sign - perf[i-1])
output[i] = lower[i] if trend[i] == 1 else upper[i]
return {
"trend": trend,
"upper": upper,
"lower": lower,
"output": output,
"perf": perf,
"factor": factor
}
factors = np.arange(min_mult, max_mult + 0.0001, step_mult)
st_results = []
for f in factors:
st_results.append(compute_supertrend(df, f, perf_alpha))
perf_vals = np.array([res["perf"][-1] for res in st_results])
fact_vals = np.array([res["factor"] for res in st_results])
# 4) K-means clustering (k=3)
def k_means(data, factors, k=3, max_iter=max_iter):
c1, c2, c3 = np.percentile(data, [25, 50, 75])
centroids = np.array([c1, c2, c3])
for _ in range(max_iter):
clusters = {0: [], 1: [], 2: []}
cluster_factors = {0: [], 1: [], 2: []}
for d, f in zip(data, factors):
dist = np.abs(d - centroids)
idx = dist.argmin()
clusters[idx].append(d)
cluster_factors[idx].append(f)
new_centroids = np.array([np.mean(clusters[i]) if len(clusters[i]) > 0 else centroids[i] for i in range(3)])
if np.allclose(new_centroids, centroids):
break
centroids = new_centroids
return clusters, cluster_factors, centroids
clusters, cluster_factors, centroids = k_means(perf_vals, fact_vals, k=3, max_iter=max_iter)
order = np.argsort(centroids)
sorted_clusters = {i: clusters[j] for i, j in enumerate(order)}
sorted_cluster_factors = {i: cluster_factors[j] for i, j in enumerate(order)}
sorted_centroids = centroids[order]
if from_cluster == "Best":
chosen_index = 2
elif from_cluster == "Average":
chosen_index = 1
else:
chosen_index = 0
if len(sorted_cluster_factors[chosen_index]) > 0:
target_factor = np.mean(sorted_cluster_factors[chosen_index])
else:
target_factor = factors[-1]
if len(sorted_clusters[chosen_index]) > 0:
target_perf = np.mean(sorted_clusters[chosen_index])
else:
target_perf = 0.0
# 5) Recompute final supertrend with target_factor
st_final = compute_supertrend(df, target_factor, perf_alpha)
ts = st_final["output"]
os_arr = np.zeros(n, dtype=int)
os_arr[0] = 1 if df["Close"].iloc[0] > st_final["upper"][0] else 0
for i in range(1, n):
c = df["Close"].iloc[i]
up = st_final["upper"][i]
dn = st_final["lower"][i]
if c > up:
os_arr[i] = 1
elif c < dn:
os_arr[i] = 0
else:
os_arr[i] = os_arr[i-1]
# Build an adaptive MA for the trailing stop
den_close_diff = (df["Close"] - df["Close"].shift(1)).abs()
den = den_close_diff.ewm(alpha=2/(perf_alpha+1), adjust=False).mean()
den_val = den.iloc[-1] if den.iloc[-1] != 0 else 1e-9
perf_idx = max(target_perf, 0) / den_val
perf_ama = np.zeros(n, dtype=float)
perf_ama[0] = ts[0]
for i in range(1, n):
perf_ama[i] = perf_ama[i-1] + perf_idx * (ts[i] - perf_ama[i-1])
# 6) Build Plotly chart
fig = go.Figure()
fig.add_trace(go.Scatter(
x=df["Date"],
y=df["Close"],
mode="lines",
line=dict(color="silver", width=1.2),
name="Close Price"
))
ts_bull = np.where(os_arr == 1, ts, np.nan)
ts_bear = np.where(os_arr == 0, ts, np.nan)
fig.add_trace(go.Scatter(
x=df["Date"],
y=ts_bull,
mode="lines",
line=dict(color="teal", width=1.2),
name="Bullish Stop"
))
fig.add_trace(go.Scatter(
x=df["Date"],
y=ts_bear,
mode="lines",
line=dict(color="red", width=1.2),
name="Bearish Stop"
))
fig.add_trace(go.Scatter(
x=df["Date"],
y=perf_ama,
mode="lines",
line=dict(color="orange", width=1.0),
opacity=0.7,
name="Trailing Stop AMA"
))
for i in range(1, n):
if os_arr[i] != os_arr[i-1]:
if os_arr[i] == 1:
fig.add_trace(go.Scatter(
x=[df["Date"].iloc[i]],
y=[ts[i]],
mode="markers",
marker=dict(symbol="triangle-up", size=10, color="teal",
line=dict(color="white", width=1)),
name="Bullish Signal",
showlegend=False
))
else:
fig.add_trace(go.Scatter(
x=[df["Date"].iloc[i]],
y=[ts[i]],
mode="markers",
marker=dict(symbol="triangle-down", size=10, color="red",
line=dict(color="white", width=1)),
name="Bearish Signal",
showlegend=False
))
fig.update_layout(
title=f"SuperTrend (Clustering) - {ticker} [Factor ~ {target_factor:.2f}]",
xaxis_title="Date",
yaxis_title="Price",
template="plotly_dark",
width=1600,
height=800
)
fig.update_xaxes(tickformat='%Y-%m-%d')
# Display results
st.markdown("### Analysis Results")
st.write("The plot below shows the closing price with the SuperTrend indicators and signals.")
st.plotly_chart(fig, use_container_width=True)
except Exception as e:
st.error("An error occurred during the analysis.")
st.error(str(e))
# Hide default Streamlit style
st.markdown(
"""
<style>
#MainMenu {visibility: hidden;}
footer {visibility: hidden;}
</style>
""",
unsafe_allow_html=True
)