---
title: "Entropic Value at Risk"
description: "Entropic Value at Risk is an allocation method minimising a tail-loss bound drawn from every day in the window: more prudent than CVaR, stable on little data."
canonical_url: "https://fincanva.com/docs/strategies/entropic-value-at-risk"
last_updated: "2026-09-24"
md_url: "https://fincanva.com/docs/strategies/entropic-value-at-risk.md"
---

# Entropic Value at Risk

Entropic Value at Risk (EVaR) is an [allocation method](/docs/strategies/allocation-and-allocation-method) that minimises a measure of tail loss built from **every** day in the window, with each day weighted more heavily the worse its loss. It sits above [CVaR](/docs/strategies/minimum-cvar) at the same tail share — always at least as prudent — and, unlike CVaR, it does not collapse onto a single worst day when the window is short. The method picker labels it **"EVaR · Entropic VaR"** and describes it as "More prudent than CVaR, stable even with little history".

**Also seen as:** EVaR, entropic VaR

## What is Entropic Value at Risk?

EVaR is the tightest upper bound on the tail loss that can be built from the whole distribution of returns rather than from its worst days alone:

$$
\text{EVaR}_\alpha(L) = \inf_{z > 0} \; z \,\ln\!\left( \frac{\mathbb{E}\left[e^{L/z}\right]}{\alpha} \right)
$$

where: $L$ is the portfolio's daily loss, $\alpha$ is the **Tail share**, $\mathbb{E}[e^{L/z}]$ averages every day's loss after an exponential weighting that grows steeply with the size of the loss, and $z$ is the scale that makes the bound tightest. In words: large losses dominate the average without the ordinary days being thrown away, so the answer depends smoothly on the weights and never on one single observation.

## How is EVaR different from Minimum CVaR?

Two differences, and both come from EVaR using every day.

- **It is more prudent.** At the same tail share, a portfolio's EVaR is never below its CVaR, so the weights that minimise it come out more conservative. That is the purpose, not a side effect.
- **It has no small-sample cliff.** [Minimum CVaR](/docs/strategies/minimum-cvar) averages the worst tail-share of days; when the window times the tail share falls below one day, it becomes a single-worst-day rule, and the app warns you. EVaR has no such limit, which is why that warning offers "Switch to EVaR" as one of its two fixes.

## How does Fincanva handle it?

- Entropic Value at Risk is offered at **both levels**: across the instruments of a strategy, and across the strategies of a [Combined](/docs/getting-started/strategy-in-a-combined).
- Its one setting is the **Tail share**, from 1% to 25% and 5% by default — the same control as Minimum CVaR.
- It reads the [calculation window](/docs/strategies/calculation-window) (**In-sample**, 12 months by default) and does not read the **Risk estimation** choice.
- No low-data warning is ever shown for it, whatever the window and tail share.
- Weights are never negative.
- The picker marks it "slow to compute": each [rebalance](/docs/backtesting/rebalance) solves an optimisation over every day of the window.

## Which plan includes Entropic Value at Risk?

It depends on your plan, at each level where the method is offered.

**Inside a strategy**

Included from Ultimate upwards. See [what each plan includes](/docs/account-security/what-each-plan-includes).

**Inside a Combined**

Included from Ultimate upwards. See [what each plan includes](/docs/account-security/what-each-plan-includes).

## What does it look like in practice?

A strategy uses a 1-month in-sample window, about 21 trading days, with a 4% tail share.

- **Minimum CVaR** would average the worst 21 × 4% = 0.84 days — less than one — so it can only look at the single worst day, and the app shows "Too little data for this tail share".
- **Entropic Value at Risk** uses all 21 days, each weighted by how bad it was. The worst day still counts most, but the second- and third-worst days move the answer too, so a small change in the data does not flip the weights.

Lengthen the window to 12 months, about 252 days, and both methods have enough data; EVaR then still reads a little more deeply into the tail than CVaR at the same 4%.

Fincanva is for education and illustration only. It is not personalised financial advice, and past or simulated results do not predict future ones. [Read the Terms Addendum](https://fincanva.com/terms/addendum#section-3)