Регистрация

Hiddenmarkovmodel

Hidden Markov Model: Regime Probability [AlgoPoint]Hidden Markov Model: Regime Probability Traditional technical indicators are deterministic and lagging; they tell you what the price has already done. The Hidden Markov Model (HMM) Regime Probability system takes a completely different, quantitative approach. It uses probabilistic mathematics to estimate the unobservable "Hidden State" (Market Regime) the price is currently operating in. Inspired by the mathematical models used by institutional quantitative hedge funds, this script doesn't just look at price direction—it calculates the probability of the market being in a specific regime based on real-time observations of Momentum and Volatility. 1. The Three Hidden States (Regimes) The market is modeled as existing in one of three hidden states: ↗ Bullish Regime: High positive momentum with low or stable volatility. (Steady, grinding uptrends). ↘ Bearish Regime: High negative momentum with high volatility. (Aggressive sell-offs and panic). ↕ Chop / Chaos Regime: Zero/low momentum with high volatility. (Whipsaw, ranging, and unpredictable noise). 2. How It Works (The Quant Engine) Since Pine Script does not natively support complex matrix optimization, this script builds a robust Pseudo-HMM using a predefined Transition Matrix and Bayesian Updates. Observables (Emissions): The script calculates the Z-Scores of Smoothed Momentum (Rate of Change) and Volatility (ATR). Emission Probabilities (Gaussian PDF): It feeds these Z-Scores into a Gaussian Probability Density Function to see how well the current market matches the expected profile of a Bull, Bear, or Chop regime. Bayesian Update: Using a predefined Markov Transition Matrix (the statistical inertia of a trend), it updates the prior probabilities to give you a real-time percentage (0-100%) for each regime. 3. Advanced Visual Features & UI We built a custom UI/UX engine to make digesting complex probabilities instantaneous: Exponential Color Smoothing (Bar Colors): As the probability of a regime increases, the bar colors smoothly transition. We implemented an exponential color blending algorithm to prevent abrupt, distracting color changes and eliminate "muddy" colors during transitions. Pro Quant Dashboard: A built-in HUD (Heads-Up Display) provides a quick summary. It features a dominant state readout, an overall "Confidence Score", and ASCII-style mini progress bars (████░░░) for rapid visual processing of probabilities without needing to read the numbers. Stacked Area Oscillator: The bottom panel displays a 0-100 stacked area chart, showing the exact distribution of probabilities across Bull (Green), Chop (Purple), and Bear (Red) states. 4. How to Use This Tool This is not a standalone Buy/Sell signal indicator. It is a Strategy Filter and a Risk Manager. When Bull/Bear Probability is Dominant (>50%): The market is trending. Turn ON your trend-following indicators (like Moving Averages or Breakout systems) and ignore overbought/oversold signals. When Chop Probability is Dominant (>50%): The market is noisy. Turn OFF your trend-following systems. Either switch to Mean Reversion strategies (like RSI or Bollinger Bands) or stay in cash until a clear regime emerges. Watch the Confidence Score: If the Dashboard shows "LOW" confidence, it means the probabilities are split (e.g., 34% Bull, 33% Chop, 33% Bear). Wait for the model to gain confidence before committing capital. 5. Alerts The script includes non-repainting alerts that trigger only when the dominant regime changes: HMM Regime: BULLISH 🚀 * HMM Regime: BEARISH 🩸 * HMM Regime: CHOP ⚖️ 6. Settings Lookback Period: The window used to calculate the Z-scores for momentum and volatility. Transition Matrix: Allows advanced users to tweak the statistical likelihood of the market staying in its current state versus transitioning to a new one. Color Transition Speed: Adjusts the smoothness of the bar coloring. A lower value creates a buttery-smooth fade between regimes, while a value of 1.0 makes it instant.
Индикатор Pine Script®
от Iamalala
Hidden Markov Model Market Regimes [LuxAlgo]The Hidden Markov Model Market Regimes tool provides a probabilistic framework for identifying the current state of the market by applying a Hidden Markov Model (HMM) to price action and volatility data. 🔶 USAGE The indicator identifies four distinct market regimes. Each regime is represented by a unique color, and the oscillator values (0-100%) represent the probability of the market being in that specific state. 🔹 Regime Breakdown & Trading Implications Low Volatility Trend (Gray): Characterized by steady, consistent returns with minimal price swings. Ideal for trend-following strategies and "buy-and-hold" positions. High Volatility Chop (Orange): Large price swings without a clear directional trend. Suggests a "risk-off" environment or mean-reversion strategies. Crash Regime (Red): Extreme negative returns coupled with a spike in volatility. Indicates high panic; traders may look for hedging opportunities. Accumulation (Blue): Low volatility with neutral-to-rounding price action. Often occurs at the end of a bear cycle, signaling a potential "bottom-fishing" zone. 🔹 Visual Cues The indicator features background highlighting that changes based on the "Dominant State" (the state with the highest current probability). When a specific probability line crosses above the 50% "Neutral" line, the model is gaining conviction. Crosses above 80% indicate "High Confidence" in that regime. Users should watch for "probability flips," where one regime's dominance is rapidly overtaken by another, signaling a structural shift in the market. 🔶 DETAILS The script implements a specialized version of the Forward Algorithm to estimate the likelihood of hidden states. 🔹 Heuristic vs. Trained ML It is important to note that this indicator is accurately described as a "Heuristic" or "Static" HMM rather than a traditional Machine Learning model. Unlike modern ML models that require extensive training on historical datasets (using algorithms like Baum-Welch or Expectation-Maximization), this model uses fixed mathematical parameters and predefined heuristic-based emission profiles. This approach ensures deterministic behavior—the model will always react to the same price patterns in the same way—and eliminates the "black box" nature of traditional ML while still providing the probabilistic benefits of a Markov chain. 🔹 Mathematical Logic Log Returns: The model utilizes log returns ( TSX:LN (Close_{t} / Close_{t-1})$) rather than simple percentage changes to ensure time-additivity and a more symmetric distribution of data. Emission Likelihoods: The model uses Gaussian-style probability density functions to compare normalized returns and volatility against "ideal" profiles for each state. For example, the Crash Regime likelihood increases when returns are significantly below the mean and volatility is significantly above the mean. Transition Matrix (Markov Property): This defines the "memory" of the system. The matrix is weighted toward "persistence" (diagonal values), meaning the market is statistically more likely to stay in its current regime than to jump to a different one every bar. 🔶 SETTINGS 🔹 HMM Settings Lookback Period: The window used to calculate the mean and standard deviation for normalizing returns and volatility. Learning Rate: Controls how quickly the model updates its internal probabilities. Range: 0.01 (very slow/stable) to 1.0 (instant/reactive). 🔹 Dashboard Enable Dashboard: Toggles the visibility of the on-screen information table. Position: Determines where the dashboard is anchored (Top Right, Bottom Right, or Bottom Left). Size: Adjusts the scale of the dashboard text and cells.
Индикатор Pine Script®
от LuxAlgo
2
Markov Chain [3D] | FractalystWhat exactly is a Markov Chain? This indicator uses a Markov Chain model to analyze, quantify, and visualize the transitions between market regimes (Bull, Bear, Neutral) on your chart. It dynamically detects these regimes in real-time, calculates transition probabilities, and displays them as animated 3D spheres and arrows, giving traders intuitive insight into current and future market conditions. How does a Markov Chain work, and how should I read this spheres-and-arrows diagram? Think of three weather modes: Sunny, Rainy, Cloudy. Each sphere is one mode. The loop on a sphere means “stay the same next step” (e.g., Sunny again tomorrow). The arrows leaving a sphere show where things usually go next if they change (e.g., Sunny moving to Cloudy). Some paths matter more than others. A more prominent loop means the current mode tends to persist. A more prominent outgoing arrow means a change to that destination is the usual next step. Direction isn’t symmetric: moving Sunny→Cloudy can behave differently than Cloudy→Sunny. Now relabel the spheres to markets: Bull, Bear, Neutral. Spheres: market regimes (uptrend, downtrend, range). Self‑loop: tendency for the current regime to continue on the next bar. Arrows: the most common next regime if a switch happens. How to read: Start at the sphere that matches current bar state. If the loop stands out, expect continuation. If one outgoing path stands out, that switch is the typical next step. Opposite directions can differ (Bear→Neutral doesn’t have to match Neutral→Bear). What states and transitions are shown? The three market states visualized are: Bullish (Bull): Upward or strong-market regime. Bearish (Bear): Downward or weak-market regime. Neutral: Sideways or range-bound regime. Bidirectional animated arrows and probability labels show how likely the market is to move from one regime to another (e.g., Bull → Bear or Neutral → Bull). How does the regime detection system work? You can use either built-in price returns (based on adaptive Z-score normalization) or supply three custom indicators (such as volume, oscillators, etc.). Values are statistically normalized (Z-scored) over a configurable lookback period. The normalized outputs are classified into Bull, Bear, or Neutral zones. If using three indicators, their regime signals are averaged and smoothed for robustness. How are transition probabilities calculated? On every confirmed bar, the algorithm tracks the sequence of detected market states, then builds a rolling window of transitions. The code maintains a transition count matrix for all regime pairs (e.g., Bull → Bear). Transition probabilities are extracted for each possible state change using Laplace smoothing for numerical stability, and frequently updated in real-time. What is unique about the visualization? 3D animated spheres represent each regime and change visually when active. Animated, bidirectional arrows reveal transition probabilities and allow you to see both dominant and less likely regime flows. Particles (moving dots) animate along the arrows, enhancing the perception of regime flow direction and speed. All elements dynamically update with each new price bar, providing a live market map in an intuitive, engaging format. Can I use custom indicators for regime classification? Yes! Enable the "Custom Indicators" switch and select any three chart series as inputs. These will be normalized and combined (each with equal weight), broadening the regime classification beyond just price-based movement. What does the “Lookback Period” control? Lookback Period (default: 100) sets how much historical data builds the probability matrix. Shorter periods adapt faster to regime changes but may be noisier. Longer periods are more stable but slower to adapt. How is this different from a Hidden Markov Model (HMM)? It sets the window for both regime detection and probability calculations. Lower values make the system more reactive, but potentially noisier. Higher values smooth estimates and make the system more robust. How is this Markov Chain different from a Hidden Markov Model (HMM)? Markov Chain (as here): All market regimes (Bull, Bear, Neutral) are directly observable on the chart. The transition matrix is built from actual detected regimes, keeping the model simple and interpretable. Hidden Markov Model: The actual regimes are unobservable ("hidden") and must be inferred from market output or indicator "emissions" using statistical learning algorithms. HMMs are more complex, can capture more subtle structure, but are harder to visualize and require additional machine learning steps for training. A standard Markov Chain models transitions between observable states using a simple transition matrix, while a Hidden Markov Model assumes the true states are hidden (latent) and must be inferred from observable “emissions” like price or volume data. In practical terms, a Markov Chain is transparent and easier to implement and interpret; an HMM is more expressive but requires statistical inference to estimate hidden states from data. Markov Chain: states are observable; you directly count or estimate transition probabilities between visible states. This makes it simpler, faster, and easier to validate and tune. HMM: states are hidden; you only observe emissions generated by those latent states. Learning involves machine learning/statistical algorithms (commonly Baum–Welch/EM for training and Viterbi for decoding) to infer both the transition dynamics and the most likely hidden state sequence from data. How does the indicator avoid “repainting” or look-ahead bias? All regime changes and matrix updates happen only on confirmed (closed) bars, so no future data is leaked, ensuring reliable real-time operation. Are there practical tuning tips? Tune the Lookback Period for your asset/timeframe: shorter for fast markets, longer for stability. Use custom indicators if your asset has unique regime drivers. Watch for rapid changes in transition probabilities as early warning of a possible regime shift. Who is this indicator for? Quants and quantitative researchers exploring probabilistic market modeling, especially those interested in regime-switching dynamics and Markov models. Programmers and system developers who need a probabilistic regime filter for systematic and algorithmic backtesting: The Markov Chain indicator is ideally suited for programmatic integration via its bias output (1 = Bull, 0 = Neutral, -1 = Bear). Although the visualization is engaging, the core output is designed for automated, rules-based workflows—not for discretionary/manual trading decisions. Developers can connect the indicator’s output directly to their Pine Script logic (using input.source()), allowing rapid and robust backtesting of regime-based strategies. It acts as a plug-and-play regime filter: simply plug the bias output into your entry/exit logic, and you have a scientifically robust, probabilistically-derived signal for filtering, timing, position sizing, or risk regimes. The MC's output is intentionally "trinary" (1/0/-1), focusing on clear regime states for unambiguous decision-making in code. If you require nuanced, multi-probability or soft-label state vectors, consider expanding the indicator or stacking it with a probability-weighted logic layer in your scripting. Because it avoids subjectivity, this approach is optimal for systematic quants, algo developers building backtested, repeatable strategies based on probabilistic regime analysis. What's the mathematical foundation behind this? The mathematical foundation behind this Markov Chain indicator—and probabilistic regime detection in finance—draws from two principal models: the (standard) Markov Chain and the Hidden Markov Model (HMM). How to use this indicator programmatically? The Markov Chain indicator automatically exports a bias value (+1 for Bullish, -1 for Bearish, 0 for Neutral) as a plot visible in the Data Window. This allows you to integrate its regime signal into your own scripts and strategies for backtesting, automation, or live trading. Step-by-Step Integration with Pine Script (input.source) Add the Markov Chain indicator to your chart. This must be done first, since your custom script will "pull" the bias signal from the indicator's plot. In your strategy, create an input using input.source() Example: //@version=5 strategy("MC Bias Strategy Example") mcBias = input.source(close, "MC Bias Source") After saving, go to your script’s settings. For the “MC Bias Source” input, select the plot/output of the Markov Chain indicator (typically its bias plot). Use the bias in your trading logic Example (long only on Bull, flat otherwise): if mcBias == 1 strategy.entry("Long", strategy.long) else strategy.close("Long") For more advanced workflows, combine mcBias with additional filters or trailing stops. How does this work behind-the-scenes? TradingView’s input.source() lets you use any plot from another indicator as a real-time, “live” data feed in your own script (source). The selected bias signal is available to your Pine code as a variable, enabling logical decisions based on regime (trend-following, mean-reversion, etc.). This enables powerful strategy modularity : decouple regime detection from entry/exit logic, allowing fast experimentation without rewriting core signal code. Integrating 45+ Indicators with Your Markov Chain — How & Why The Enhanced Custom Indicators Export script exports a massive suite of over 45 technical indicators—ranging from classic momentum (RSI, MACD, Stochastic, etc.) to trend, volume, volatility, and oscillator tools—all pre-calculated, centered/scaled, and available as plots. // Enhanced Custom Indicators Export - 45 Technical Indicators // Comprehensive technical analysis suite for advanced market regime detection //@version=6 indicator('Enhanced Custom Indicators Export | Fractalyst', shorttitle='Enhanced CI Export', overlay=false, scale=scale.right, max_labels_count=500, max_lines_count=500) // |----- Input Parameters -----| // momentum_group = "Momentum Indicators" trend_group = "Trend Indicators" volume_group = "Volume Indicators" volatility_group = "Volatility Indicators" oscillator_group = "Oscillator Indicators" display_group = "Display Settings" // Common lengths length_14 = input.int(14, "Standard Length (14)", minval=1, maxval=100, group=momentum_group) length_20 = input.int(20, "Medium Length (20)", minval=1, maxval=200, group=trend_group) length_50 = input.int(50, "Long Length (50)", minval=1, maxval=200, group=trend_group) // Display options show_table = input.bool(true, "Show Values Table", group=display_group) table_size = input.string("Small", "Table Size", options= , group=display_group) // |----- MOMENTUM INDICATORS (15 indicators) -----| // // 1. RSI (Relative Strength Index) rsi_14 = ta.rsi(close, length_14) rsi_centered = rsi_14 - 50 // 2. Stochastic Oscillator stoch_k = ta.stoch(close, high, low, length_14) stoch_d = ta.sma(stoch_k, 3) stoch_centered = stoch_k - 50 // 3. Williams %R williams_r = ta.stoch(close, high, low, length_14) - 100 // 4. MACD (Moving Average Convergence Divergence) = ta.macd(close, 12, 26, 9) // 5. Momentum (Rate of Change) momentum = ta.mom(close, length_14) momentum_pct = (momentum / close ) * 100 // 6. Rate of Change (ROC) roc = ta.roc(close, length_14) // 7. Commodity Channel Index (CCI) cci = ta.cci(close, length_20) // 8. Money Flow Index (MFI) mfi = ta.mfi(close, length_14) mfi_centered = mfi - 50 // 9. Awesome Oscillator (AO) ao = ta.sma(hl2, 5) - ta.sma(hl2, 34) // 10. Accelerator Oscillator (AC) ac = ao - ta.sma(ao, 5) // 11. Chande Momentum Oscillator (CMO) cmo = ta.cmo(close, length_14) // 12. Detrended Price Oscillator (DPO) dpo = close - ta.sma(close, length_20) // 13. Price Oscillator (PPO) ppo = ta.sma(close, 12) - ta.sma(close, 26) ppo_pct = (ppo / ta.sma(close, 26)) * 100 // 14. TRIX trix_ema1 = ta.ema(close, length_14) trix_ema2 = ta.ema(trix_ema1, length_14) trix_ema3 = ta.ema(trix_ema2, length_14) trix = ta.roc(trix_ema3, 1) * 10000 // 15. Klinger Oscillator klinger = ta.ema(volume * (high + low + close) / 3, 34) - ta.ema(volume * (high + low + close) / 3, 55) // 16. Fisher Transform fisher_hl2 = 0.5 * (hl2 - ta.lowest(hl2, 10)) / (ta.highest(hl2, 10) - ta.lowest(hl2, 10)) - 0.25 fisher = 0.5 * math.log((1 + fisher_hl2) / (1 - fisher_hl2)) // 17. Stochastic RSI stoch_rsi = ta.stoch(rsi_14, rsi_14, rsi_14, length_14) stoch_rsi_centered = stoch_rsi - 50 // 18. Relative Vigor Index (RVI) rvi_num = ta.swma(close - open) rvi_den = ta.swma(high - low) rvi = rvi_den != 0 ? rvi_num / rvi_den : 0 // 19. Balance of Power (BOP) bop = (close - open) / (high - low) // |----- TREND INDICATORS (10 indicators) -----| // // 20. Simple Moving Average Momentum sma_20 = ta.sma(close, length_20) sma_momentum = ((close - sma_20) / sma_20) * 100 // 21. Exponential Moving Average Momentum ema_20 = ta.ema(close, length_20) ema_momentum = ((close - ema_20) / ema_20) * 100 // 22. Parabolic SAR sar = ta.sar(0.02, 0.02, 0.2) sar_trend = close > sar ? 1 : -1 // 23. Linear Regression Slope lr_slope = ta.linreg(close, length_20, 0) - ta.linreg(close, length_20, 1) // 24. Moving Average Convergence (MAC) mac = ta.sma(close, 10) - ta.sma(close, 30) // 25. Trend Intensity Index (TII) tii_sum = 0.0 for i = 1 to length_20 tii_sum += close > close ? 1 : 0 tii = (tii_sum / length_20) * 100 // 26. Ichimoku Cloud Components ichimoku_tenkan = (ta.highest(high, 9) + ta.lowest(low, 9)) / 2 ichimoku_kijun = (ta.highest(high, 26) + ta.lowest(low, 26)) / 2 ichimoku_signal = ichimoku_tenkan > ichimoku_kijun ? 1 : -1 // 27. MESA Adaptive Moving Average (MAMA) mama_alpha = 2.0 / (length_20 + 1) mama = ta.ema(close, length_20) mama_momentum = ((close - mama) / mama) * 100 // 28. Zero Lag Exponential Moving Average (ZLEMA) zlema_lag = math.round((length_20 - 1) / 2) zlema_data = close + (close - close ) zlema = ta.ema(zlema_data, length_20) zlema_momentum = ((close - zlema) / zlema) * 100 // |----- VOLUME INDICATORS (6 indicators) -----| // // 29. On-Balance Volume (OBV) obv = ta.obv // 30. Volume Rate of Change (VROC) vroc = ta.roc(volume, length_14) // 31. Price Volume Trend (PVT) pvt = ta.pvt // 32. Negative Volume Index (NVI) nvi = 0.0 nvi := volume < volume ? nvi + ((close - close ) / close ) * nvi : nvi // 33. Positive Volume Index (PVI) pvi = 0.0 pvi := volume > volume ? pvi + ((close - close ) / close ) * pvi : pvi // 34. Volume Oscillator vol_osc = ta.sma(volume, 5) - ta.sma(volume, 10) // 35. Ease of Movement (EOM) eom_distance = high - low eom_box_height = volume / 1000000 eom = eom_box_height != 0 ? eom_distance / eom_box_height : 0 eom_sma = ta.sma(eom, length_14) // 36. Force Index force_index = volume * (close - close ) force_index_sma = ta.sma(force_index, length_14) // |----- VOLATILITY INDICATORS (10 indicators) -----| // // 37. Average True Range (ATR) atr = ta.atr(length_14) atr_pct = (atr / close) * 100 // 38. Bollinger Bands Position bb_basis = ta.sma(close, length_20) bb_dev = 2.0 * ta.stdev(close, length_20) bb_upper = bb_basis + bb_dev bb_lower = bb_basis - bb_dev bb_position = bb_dev != 0 ? (close - bb_basis) / bb_dev : 0 bb_width = bb_dev != 0 ? (bb_upper - bb_lower) / bb_basis * 100 : 0 // 39. Keltner Channels Position kc_basis = ta.ema(close, length_20) kc_range = ta.ema(ta.tr, length_20) kc_upper = kc_basis + (2.0 * kc_range) kc_lower = kc_basis - (2.0 * kc_range) kc_position = kc_range != 0 ? (close - kc_basis) / kc_range : 0 // 40. Donchian Channels Position dc_upper = ta.highest(high, length_20) dc_lower = ta.lowest(low, length_20) dc_basis = (dc_upper + dc_lower) / 2 dc_position = (dc_upper - dc_lower) != 0 ? (close - dc_basis) / (dc_upper - dc_lower) : 0 // 41. Standard Deviation std_dev = ta.stdev(close, length_20) std_dev_pct = (std_dev / close) * 100 // 42. Relative Volatility Index (RVI) rvi_up = ta.stdev(close > close ? close : 0, length_14) rvi_down = ta.stdev(close < close ? close : 0, length_14) rvi_total = rvi_up + rvi_down rvi_volatility = rvi_total != 0 ? (rvi_up / rvi_total) * 100 : 50 // 43. Historical Volatility hv_returns = math.log(close / close ) hv = ta.stdev(hv_returns, length_20) * math.sqrt(252) * 100 // 44. Garman-Klass Volatility gk_vol = math.log(high/low) * math.log(high/low) - (2*math.log(2)-1) * math.log(close/open) * math.log(close/open) gk_volatility = math.sqrt(ta.sma(gk_vol, length_20)) * 100 // 45. Parkinson Volatility park_vol = math.log(high/low) * math.log(high/low) parkinson = math.sqrt(ta.sma(park_vol, length_20) / (4 * math.log(2))) * 100 // 46. Rogers-Satchell Volatility rs_vol = math.log(high/close) * math.log(high/open) + math.log(low/close) * math.log(low/open) rogers_satchell = math.sqrt(ta.sma(rs_vol, length_20)) * 100 // |----- OSCILLATOR INDICATORS (5 indicators) -----| // // 47. Elder Ray Index elder_bull = high - ta.ema(close, 13) elder_bear = low - ta.ema(close, 13) elder_power = elder_bull + elder_bear // 48. Schaff Trend Cycle (STC) stc_macd = ta.ema(close, 23) - ta.ema(close, 50) stc_k = ta.stoch(stc_macd, stc_macd, stc_macd, 10) stc_d = ta.ema(stc_k, 3) stc = ta.stoch(stc_d, stc_d, stc_d, 10) // 49. Coppock Curve coppock_roc1 = ta.roc(close, 14) coppock_roc2 = ta.roc(close, 11) coppock = ta.wma(coppock_roc1 + coppock_roc2, 10) // 50. Know Sure Thing (KST) kst_roc1 = ta.roc(close, 10) kst_roc2 = ta.roc(close, 15) kst_roc3 = ta.roc(close, 20) kst_roc4 = ta.roc(close, 30) kst = ta.sma(kst_roc1, 10) + 2*ta.sma(kst_roc2, 10) + 3*ta.sma(kst_roc3, 10) + 4*ta.sma(kst_roc4, 15) // 51. Percentage Price Oscillator (PPO) ppo_line = ((ta.ema(close, 12) - ta.ema(close, 26)) / ta.ema(close, 26)) * 100 ppo_signal = ta.ema(ppo_line, 9) ppo_histogram = ppo_line - ppo_signal // |----- PLOT MAIN INDICATORS -----| // // Plot key momentum indicators plot(rsi_centered, title="01_RSI_Centered", color=color.purple, linewidth=1) plot(stoch_centered, title="02_Stoch_Centered", color=color.blue, linewidth=1) plot(williams_r, title="03_Williams_R", color=color.red, linewidth=1) plot(macd_histogram, title="04_MACD_Histogram", color=color.orange, linewidth=1) plot(cci, title="05_CCI", color=color.green, linewidth=1) // Plot trend indicators plot(sma_momentum, title="06_SMA_Momentum", color=color.navy, linewidth=1) plot(ema_momentum, title="07_EMA_Momentum", color=color.maroon, linewidth=1) plot(sar_trend, title="08_SAR_Trend", color=color.teal, linewidth=1) plot(lr_slope, title="09_LR_Slope", color=color.lime, linewidth=1) plot(mac, title="10_MAC", color=color.fuchsia, linewidth=1) // Plot volatility indicators plot(atr_pct, title="11_ATR_Pct", color=color.yellow, linewidth=1) plot(bb_position, title="12_BB_Position", color=color.aqua, linewidth=1) plot(kc_position, title="13_KC_Position", color=color.olive, linewidth=1) plot(std_dev_pct, title="14_StdDev_Pct", color=color.silver, linewidth=1) plot(bb_width, title="15_BB_Width", color=color.gray, linewidth=1) // Plot volume indicators plot(vroc, title="16_VROC", color=color.blue, linewidth=1) plot(eom_sma, title="17_EOM", color=color.red, linewidth=1) plot(vol_osc, title="18_Vol_Osc", color=color.green, linewidth=1) plot(force_index_sma, title="19_Force_Index", color=color.orange, linewidth=1) plot(obv, title="20_OBV", color=color.purple, linewidth=1) // Plot additional oscillators plot(ao, title="21_Awesome_Osc", color=color.navy, linewidth=1) plot(cmo, title="22_CMO", color=color.maroon, linewidth=1) plot(dpo, title="23_DPO", color=color.teal, linewidth=1) plot(trix, title="24_TRIX", color=color.lime, linewidth=1) plot(fisher, title="25_Fisher", color=color.fuchsia, linewidth=1) // Plot more momentum indicators plot(mfi_centered, title="26_MFI_Centered", color=color.yellow, linewidth=1) plot(ac, title="27_AC", color=color.aqua, linewidth=1) plot(ppo_pct, title="28_PPO_Pct", color=color.olive, linewidth=1) plot(stoch_rsi_centered, title="29_StochRSI_Centered", color=color.silver, linewidth=1) plot(klinger, title="30_Klinger", color=color.gray, linewidth=1) // Plot trend continuation plot(tii, title="31_TII", color=color.blue, linewidth=1) plot(ichimoku_signal, title="32_Ichimoku_Signal", color=color.red, linewidth=1) plot(mama_momentum, title="33_MAMA_Momentum", color=color.green, linewidth=1) plot(zlema_momentum, title="34_ZLEMA_Momentum", color=color.orange, linewidth=1) plot(bop, title="35_BOP", color=color.purple, linewidth=1) // Plot volume continuation plot(nvi, title="36_NVI", color=color.navy, linewidth=1) plot(pvi, title="37_PVI", color=color.maroon, linewidth=1) plot(momentum_pct, title="38_Momentum_Pct", color=color.teal, linewidth=1) plot(roc, title="39_ROC", color=color.lime, linewidth=1) plot(rvi, title="40_RVI", color=color.fuchsia, linewidth=1) // Plot volatility continuation plot(dc_position, title="41_DC_Position", color=color.yellow, linewidth=1) plot(rvi_volatility, title="42_RVI_Volatility", color=color.aqua, linewidth=1) plot(hv, title="43_Historical_Vol", color=color.olive, linewidth=1) plot(gk_volatility, title="44_GK_Volatility", color=color.silver, linewidth=1) plot(parkinson, title="45_Parkinson_Vol", color=color.gray, linewidth=1) // Plot final oscillators plot(rogers_satchell, title="46_RS_Volatility", color=color.blue, linewidth=1) plot(elder_power, title="47_Elder_Power", color=color.red, linewidth=1) plot(stc, title="48_STC", color=color.green, linewidth=1) plot(coppock, title="49_Coppock", color=color.orange, linewidth=1) plot(kst, title="50_KST", color=color.purple, linewidth=1) // Plot final indicators plot(ppo_histogram, title="51_PPO_Histogram", color=color.navy, linewidth=1) plot(pvt, title="52_PVT", color=color.maroon, linewidth=1) // |----- Reference Lines -----| // hline(0, "Zero Line", color=color.gray, linestyle=hline.style_dashed, linewidth=1) hline(50, "Midline", color=color.gray, linestyle=hline.style_dotted, linewidth=1) hline(-50, "Lower Midline", color=color.gray, linestyle=hline.style_dotted, linewidth=1) hline(25, "Upper Threshold", color=color.gray, linestyle=hline.style_dotted, linewidth=1) hline(-25, "Lower Threshold", color=color.gray, linestyle=hline.style_dotted, linewidth=1) // |----- Enhanced Information Table -----| // if show_table and barstate.islast table_position = position.top_right table_text_size = table_size == "Tiny" ? size.tiny : table_size == "Small" ? size.small : size.normal var table info_table = table.new(table_position, 3, 18, bgcolor=color.new(color.white, 85), border_width=1, border_color=color.gray) // Headers table.cell(info_table, 0, 0, 'Category', text_color=color.black, text_size=table_text_size, bgcolor=color.new(color.blue, 70)) table.cell(info_table, 1, 0, 'Indicator', text_color=color.black, text_size=table_text_size, bgcolor=color.new(color.blue, 70)) table.cell(info_table, 2, 0, 'Value', text_color=color.black, text_size=table_text_size, bgcolor=color.new(color.blue, 70)) // Key Momentum Indicators table.cell(info_table, 0, 1, 'MOMENTUM', text_color=color.purple, text_size=table_text_size, bgcolor=color.new(color.purple, 90)) table.cell(info_table, 1, 1, 'RSI Centered', text_color=color.purple, text_size=table_text_size) table.cell(info_table, 2, 1, str.tostring(rsi_centered, '0.00'), text_color=color.purple, text_size=table_text_size) table.cell(info_table, 0, 2, '', text_color=color.blue, text_size=table_text_size) table.cell(info_table, 1, 2, 'Stoch Centered', text_color=color.blue, text_size=table_text_size) table.cell(info_table, 2, 2, str.tostring(stoch_centered, '0.00'), text_color=color.blue, text_size=table_text_size) table.cell(info_table, 0, 3, '', text_color=color.red, text_size=table_text_size) table.cell(info_table, 1, 3, 'Williams %R', text_color=color.red, text_size=table_text_size) table.cell(info_table, 2, 3, str.tostring(williams_r, '0.00'), text_color=color.red, text_size=table_text_size) table.cell(info_table, 0, 4, '', text_color=color.orange, text_size=table_text_size) table.cell(info_table, 1, 4, 'MACD Histogram', text_color=color.orange, text_size=table_text_size) table.cell(info_table, 2, 4, str.tostring(macd_histogram, '0.000'), text_color=color.orange, text_size=table_text_size) table.cell(info_table, 0, 5, '', text_color=color.green, text_size=table_text_size) table.cell(info_table, 1, 5, 'CCI', text_color=color.green, text_size=table_text_size) table.cell(info_table, 2, 5, str.tostring(cci, '0.00'), text_color=color.green, text_size=table_text_size) // Key Trend Indicators table.cell(info_table, 0, 6, 'TREND', text_color=color.navy, text_size=table_text_size, bgcolor=color.new(color.navy, 90)) table.cell(info_table, 1, 6, 'SMA Momentum %', text_color=color.navy, text_size=table_text_size) table.cell(info_table, 2, 6, str.tostring(sma_momentum, '0.00'), text_color=color.navy, text_size=table_text_size) table.cell(info_table, 0, 7, '', text_color=color.maroon, text_size=table_text_size) table.cell(info_table, 1, 7, 'EMA Momentum %', text_color=color.maroon, text_size=table_text_size) table.cell(info_table, 2, 7, str.tostring(ema_momentum, '0.00'), text_color=color.maroon, text_size=table_text_size) table.cell(info_table, 0, 8, '', text_color=color.teal, text_size=table_text_size) table.cell(info_table, 1, 8, 'SAR Trend', text_color=color.teal, text_size=table_text_size) table.cell(info_table, 2, 8, str.tostring(sar_trend, '0'), text_color=color.teal, text_size=table_text_size) table.cell(info_table, 0, 9, '', text_color=color.lime, text_size=table_text_size) table.cell(info_table, 1, 9, 'Linear Regression', text_color=color.lime, text_size=table_text_size) table.cell(info_table, 2, 9, str.tostring(lr_slope, '0.000'), text_color=color.lime, text_size=table_text_size) // Key Volatility Indicators table.cell(info_table, 0, 10, 'VOLATILITY', text_color=color.yellow, text_size=table_text_size, bgcolor=color.new(color.yellow, 90)) table.cell(info_table, 1, 10, 'ATR %', text_color=color.yellow, text_size=table_text_size) table.cell(info_table, 2, 10, str.tostring(atr_pct, '0.00'), text_color=color.yellow, text_size=table_text_size) table.cell(info_table, 0, 11, '', text_color=color.aqua, text_size=table_text_size) table.cell(info_table, 1, 11, 'BB Position', text_color=color.aqua, text_size=table_text_size) table.cell(info_table, 2, 11, str.tostring(bb_position, '0.00'), text_color=color.aqua, text_size=table_text_size) table.cell(info_table, 0, 12, '', text_color=color.olive, text_size=table_text_size) table.cell(info_table, 1, 12, 'KC Position', text_color=color.olive, text_size=table_text_size) table.cell(info_table, 2, 12, str.tostring(kc_position, '0.00'), text_color=color.olive, text_size=table_text_size) // Key Volume Indicators table.cell(info_table, 0, 13, 'VOLUME', text_color=color.blue, text_size=table_text_size, bgcolor=color.new(color.blue, 90)) table.cell(info_table, 1, 13, 'Volume ROC', text_color=color.blue, text_size=table_text_size) table.cell(info_table, 2, 13, str.tostring(vroc, '0.00'), text_color=color.blue, text_size=table_text_size) table.cell(info_table, 0, 14, '', text_color=color.red, text_size=table_text_size) table.cell(info_table, 1, 14, 'EOM', text_color=color.red, text_size=table_text_size) table.cell(info_table, 2, 14, str.tostring(eom_sma, '0.000'), text_color=color.red, text_size=table_text_size) // Key Oscillators table.cell(info_table, 0, 15, 'OSCILLATORS', text_color=color.purple, text_size=table_text_size, bgcolor=color.new(color.purple, 90)) table.cell(info_table, 1, 15, 'Awesome Osc', text_color=color.blue, text_size=table_text_size) table.cell(info_table, 2, 15, str.tostring(ao, '0.000'), text_color=color.blue, text_size=table_text_size) table.cell(info_table, 0, 16, '', text_color=color.red, text_size=table_text_size) table.cell(info_table, 1, 16, 'Fisher Transform', text_color=color.red, text_size=table_text_size) table.cell(info_table, 2, 16, str.tostring(fisher, '0.000'), text_color=color.red, text_size=table_text_size) // Summary Statistics table.cell(info_table, 0, 17, 'SUMMARY', text_color=color.black, text_size=table_text_size, bgcolor=color.new(color.gray, 70)) table.cell(info_table, 1, 17, 'Total Indicators: 52', text_color=color.black, text_size=table_text_size) regime_color = rsi_centered > 10 ? color.green : rsi_centered < -10 ? color.red : color.gray regime_text = rsi_centered > 10 ? "BULLISH" : rsi_centered < -10 ? "BEARISH" : "NEUTRAL" table.cell(info_table, 2, 17, regime_text, text_color=regime_color, text_size=table_text_size) This makes it the perfect “indicator backbone” for quantitative and systematic traders who want to prototype, combine, and test new regime detection models—especially in combination with the Markov Chain indicator. How to use this script with the Markov Chain for research and backtesting: Add the Enhanced Indicator Export to your chart. Every calculated indicator is available as an individual data stream. Connect the indicator(s) you want as custom input(s) to the Markov Chain’s “Custom Indicators” option. In the Markov Chain indicator’s settings, turn ON the custom indicator mode. For each of the three custom indicator inputs, select the exported plot from the Enhanced Export script—the menu lists all 45+ signals by name. This creates a powerful, modular regime-detection engine where you can mix-and-match momentum, trend, volume, or custom combinations for advanced filtering. Backtest regime logic directly. Once you’ve connected your chosen indicators, the Markov Chain script performs regime detection (Bull/Neutral/Bear) based on your selected features—not just price returns. The regime detection is robust, automatically normalized (using Z-score), and outputs bias (1, -1, 0) for plug-and-play integration. Export the regime bias for programmatic use. As described above, use input.source() in your Pine Script strategy or system and link the bias output. You can now filter signals, control trade direction/size, or design pairs-trading that respect true, indicator-driven market regimes. With this framework, you’re not limited to static or simplistic regime filters. You can rigorously define, test, and refine what “market regime” means for your strategies—using the technical features that matter most to you. Optimize your signal generation by backtesting across a universe of meaningful indicator blends. Enhance risk management with objective, real-time regime boundaries. Accelerate your research: iterate quickly, swap indicator components, and see results with minimal code changes. Automate multi-asset or pairs-trading by integrating regime context directly into strategy logic. Add both scripts to your chart, connect your preferred features, and start investigating your best regime-based trades—entirely within the TradingView ecosystem. References & Further Reading Ang, A., & Bekaert, G. (2002). “Regime Switches in Interest Rates.” Journal of Business & Economic Statistics, 20(2), 163–182. Hamilton, J. D. (1989). “A New Approach to the Economic Analysis of Nonstationary Time Series and the Business Cycle.” Econometrica, 57(2), 357–384. Markov, A. A. (1906). "Extension of the Limit Theorems of Probability Theory to a Sum of Variables Connected in a Chain." The Notes of the Imperial Academy of Sciences of St. Petersburg. Guidolin, M., & Timmermann, A. (2007). “Asset Allocation under Multivariate Regime Switching.” Journal of Economic Dynamics and Control, 31(11), 3503–3544. Murphy, J. J. (1999). Technical Analysis of the Financial Markets. New York Institute of Finance. Brock, W., Lakonishok, J., & LeBaron, B. (1992). “Simple Technical Trading Rules and the Stochastic Properties of Stock Returns.” Journal of Finance, 47(5), 1731–1764. Zucchini, W., MacDonald, I. L., & Langrock, R. (2017). Hidden Markov Models for Time Series: An Introduction Using R (2nd ed.). Chapman and Hall/CRC. On Quantitative Finance and Markov Models: Lo, A. W., & Hasanhodzic, J. (2009). The Heretics of Finance: Conversations with Leading Practitioners of Technical Analysis. Bloomberg Press. Patterson, S. (2016). The Man Who Solved the Market: How Jim Simons Launched the Quant Revolution. Penguin Press. TradingView Pine Script Documentation: www.tradingview.com TradingView Blog: “Use an Input From Another Indicator With Your Strategy” www.tradingview.com GeeksforGeeks: “What is the Difference Between Markov Chains and Hidden Markov Models?” www.geeksforgeeks.org What makes this indicator original and unique? - On‑chart, real‑time Markov. The chain is drawn directly on your chart. You see the current regime, its tendency to stay (self‑loop), and the usual next step (arrows) as bars confirm. - Source‑agnostic by design. The engine runs on any series you select via input.source() — price, your own oscillator, a composite score, anything you compute in the script. - Automatic normalization + regime mapping. Different inputs live on different scales. The script standardizes your chosen source and maps it into clear regimes (e.g., Bull / Bear / Neutral) without you micromanaging thresholds each time. - Rolling, bar‑by‑bar learning. Transition tendencies are computed from a rolling window of confirmed bars. What you see is exactly what the market did in that window. - Fast experimentation. Switch the source, adjust the window, and the Markov view updates instantly. It’s a rapid way to test ideas and feel regime persistence/switch behavior. Integrate your own signals (using input.source()) - In settings, choose the Source . This is powered by input.source() . - Feed it price, an indicator you compute inside the script, or a custom composite series. - The script will automatically normalize that series and process it through the Markov engine, mapping it to regimes and updating the on‑chart spheres/arrows in real time. Credits: Deep gratitude to @RicardoSantos for both the foundational Markov chain processing engine and inspiring open-source contributions, which made advanced probabilistic market modeling accessible to the TradingView community. Special thanks to @Alien_Algorithms for the innovative and visually stunning 3D sphere logic that powers the indicator’s animated, regime-based visualization. Disclaimer This tool summarizes recent behavior. It is not financial advice and not a guarantee of future results.
Индикатор Pine Script®
от Fractalyst
10