OPEN-SOURCE SCRIPT
Markov Regime Detector - CriptoMatiko
The **Markov Regime Detector** classifies each bar as Bull, Bear, or Lateral using rolling log-return, then builds a 3×3 Markov transition matrix from the chart's full visible history.
The matrix diagonal measures **persistence** — the probability of each regime continuing into the next bar. Alerts fire only when persistence exceeds a user-defined threshold, filtering out short-lived, low-confidence regime flips.
---
**How it works**
1. Computes log-return over the last N bars (default: 20)
2. Labels each bar: Bull (log-ret > +5%), Bear (log-ret < −5%), Lateral (in between)
3. Counts all historical transitions and row-normalizes into probabilities
4. Diagonal of the matrix = persistence of the current regime
5. Alert fires when: regime changes AND persistence >= configured threshold
---
**Visuals**
- Ribbon — real-time background tint: green (Bull) / rose (Bear) / grey (Lateral)
- Triangles ▲▼◆ — mark the start of each confirmed new regime
- Floating labels — regime name + persistence % at each transition point
- Table (bottom-right) — current regime, persistence, log-return and alert status
Note: The ribbon updates on the forming bar. Labels, shapes and alerts only fire on confirmed bar close (barstate.isconfirmed).
---
**Alerts**
Set up as "Any alert() call" + "Once per bar close". Dynamic message format:
REGIME: BEAR
Anterior: LATERAL
Persistencia: 85%
Preco: 2022.60
Indicador: Markov | CriptoMatiko
---
**Parameters**
Lookback window (default 20): bars used to compute log-return and classify the current state
Bull threshold (default 5%): log-ret above this value = Bull
Bear threshold (default 5%): log-ret below -5% = Bear
Min persistence to alert (default 55%): minimum persistence to trigger an alert
Min bars to hold (default 3): consecutive bars in same regime before label appears
---
The matrix diagonal measures **persistence** — the probability of each regime continuing into the next bar. Alerts fire only when persistence exceeds a user-defined threshold, filtering out short-lived, low-confidence regime flips.
---
**How it works**
1. Computes log-return over the last N bars (default: 20)
2. Labels each bar: Bull (log-ret > +5%), Bear (log-ret < −5%), Lateral (in between)
3. Counts all historical transitions and row-normalizes into probabilities
4. Diagonal of the matrix = persistence of the current regime
5. Alert fires when: regime changes AND persistence >= configured threshold
---
**Visuals**
- Ribbon — real-time background tint: green (Bull) / rose (Bear) / grey (Lateral)
- Triangles ▲▼◆ — mark the start of each confirmed new regime
- Floating labels — regime name + persistence % at each transition point
- Table (bottom-right) — current regime, persistence, log-return and alert status
Note: The ribbon updates on the forming bar. Labels, shapes and alerts only fire on confirmed bar close (barstate.isconfirmed).
---
**Alerts**
Set up as "Any alert() call" + "Once per bar close". Dynamic message format:
REGIME: BEAR
Anterior: LATERAL
Persistencia: 85%
Preco: 2022.60
Indicador: Markov | CriptoMatiko
---
**Parameters**
Lookback window (default 20): bars used to compute log-return and classify the current state
Bull threshold (default 5%): log-ret above this value = Bull
Bear threshold (default 5%): log-ret below -5% = Bear
Min persistence to alert (default 55%): minimum persistence to trigger an alert
Min bars to hold (default 3): consecutive bars in same regime before label appears
---
Open-source script
In true TradingView spirit, the creator of this script has made it open-source, so that traders can review and verify its functionality. Kudos to the author! While you can use it for free, remember that republishing the code is subject to our House Rules.
Disclaimer
The information and publications are not meant to be, and do not constitute, financial, investment, trading, or other types of advice or recommendations supplied or endorsed by TradingView. Read more in the Terms of Use.
Open-source script
In true TradingView spirit, the creator of this script has made it open-source, so that traders can review and verify its functionality. Kudos to the author! While you can use it for free, remember that republishing the code is subject to our House Rules.
Disclaimer
The information and publications are not meant to be, and do not constitute, financial, investment, trading, or other types of advice or recommendations supplied or endorsed by TradingView. Read more in the Terms of Use.