---
title: Minimum-Ratio Estimator
url: https://www.emergentmind.com/topics/minimum-ratio-estimator
type: topic
---

# Minimum-Ratio Estimator

A minimum-ratio estimator is a statistical construction designed to estimate a population parameter expressed as a ratio (e.g., prevalence, relative risk, odds ratio), with properties of minimax optimality, control over mean-square error, and—in some cases—guaranteed efficiency relative to the Cramér–Rao lower bound. These estimators appear in contexts such as quantification under prior-probability shift, finite-population inference with auxiliary variables, and design-based estimation of relative risk with controlled precision and resource allocation. The methodology is unified by the use of ratio forms and optimization over risk or mean-square error subject to specified constraints.

## 1. Ratio Estimators Under Prior-Probability Shift

When only a labeled source sample and an unlabeled target sample are available, quantification under prior-probability shift seeks estimation of the target prevalence $\pi_t = P_{\mathrm{target}}(Y=1)$. The prior-shift assumption, $P_{\mathrm{train}}(X|Y) \equiv P_{\mathrm{target}}(X|Y)$, entails that the mixture distribution of features in the target is a convex combination of class-conditional densities $f_0(x),f_1(x)$ with mixture weights $\pi_t$.

The ratio estimator is constructed as:
\[
\hat \pi_{UR} = \frac{\mu_t - \mu_0^s}{\mu_1^s - \mu_0^s}, \qquad
\mu_0^s = \frac{1}{n_{s,0}} \sum_{i:Y_i=0} g(X_i),\;\; \mu_1^s = \frac{1}{n_{s,1}} \sum_{i:Y_i=1} g(X_i),\;\;\mu_t = \frac{1}{n_t} \sum_{j=1}^{n_t} g(X_j)
\]
for a scoring function $g$ satisfying $\Delta = \mathbb{E}[g(X)|Y=1]-\mathbb{E}[g(X)|Y=0] \ne 0$. This estimator achieves approximate minimax optimality in $\ell_2$-risk over separable parameter classes, with the worst-case risk scaling as $O(\max(n_s^{-1}, n_t^{-1}))$, matching the established lower bound. Trimmed forms $\hat \pi_R = \min\{1,\max\{0,\hat \pi_{UR}\}\}$ are standard.

With additional knowledge of class-conditional densities, the estimator can be recast using importance weights $w(x)=f_1(x)/f_0(x)$:
\[
\pi_t = \frac{\frac{1}{n_t}\sum_{j} w(X_j) - 1}{\theta^{-1} - 1},\;\;\theta = \mathbb{E}[w(X)|Y=1]^{-1}
\]
Several classical approaches (adjusted count, EM prior-shift correction) are nested in this framework [1807.03929].

## 2. Minimum-Ratio Estimation With Controlled Relative Error

In two-sample binary experiments, the minimum-ratio estimator constructs the estimate of relative risk $R=p_1/p_2$ ensuring the relative mean-square error (RMSE) is bounded by a target $\varepsilon^2$ for all $p_1,p_2\in(0,1)$, and average sample sizes are in a user-specified ratio $\rho$. This is achieved by a two-stage inverse binomial sampling protocol:

- **Stage I (pilot):** For each population $i=1,2$, sample until $r_i$ successes; record number of draws $M_i$. Compute preliminary ratio $X = (M_2 - \tfrac12)/(M_1 - \tfrac12)$.
- **Stage II (main):** Determine $(s_1, s_2)$ to satisfy
  \[
  e(s_1,s_2) = \frac{1}{s_1-2} + \frac{1}{s_2} + \frac{1}{(s_1-2)s_2} = \varepsilon^2,\;\; \frac{r_1 + s_1}{r_2 + s_2} = \rho\;
  \]
  and sample until $s_i$ further successes (per population). The minimum-ratio (RR) estimator is then
  \[
  \hat R = \frac{s_1-1}{s_2} \frac{N_2}{N_1-1}
  \]
  where $N_i$ is the number of draws to reach $s_i$ successes in population $i$.

This estimator is unbiased, and achieves $\mathbb{E}[(\hat R - R)^2]/R^2 < \varepsilon^2$. For small $\varepsilon$, its efficiency approaches the Cramér–Rao lower bound. Batch (group) sampling operates identically, with cost adjustments for incomplete batch use [2503.04876].

## 3. Ratio-Product Estimators With Auxiliary Information

For finite populations, the two-parameter ratio-product-ratio estimator uses both sample and population information from auxiliary variable $x_i$ to estimate the mean of study variable $y_i$:
\[
\hat Y_{\alpha,\beta} = \alpha \left[\frac{(1-\beta)\bar x + \beta X}{\beta \bar x + (1-\beta) X}\right]\bar y + (1-\alpha)\left[\frac{\beta \bar x + (1-\beta) X}{(1-\beta)\bar x + \beta X}\right]\bar y
\]
Here $\alpha,\beta\in\mathbb{R}$ are tuning parameters; $\bar y, \bar x$ denote sample means, $X$ the known population mean of $x_i$. Specializations yield the sample mean, classical ratio, and product estimators for certain $(\alpha,\beta)$.

A first-order expansion yields bias and MSE expressions in terms of population variances $S_Y^2$, $S_X^2$, and correlation $\rho$. The MSE is minimized along the curve $(1-2\alpha)(1-2\beta) = \rho$, yielding the minimum-MSE estimator:
\[
\min_{\alpha,\beta} \operatorname{MSE}_1(\hat Y_{\alpha,\beta}) = \frac{1-f}{n} S_Y^2 (1-\rho^2)
\]
For strongly correlated auxiliary information, empirical results show the minimum-ratio estimator significantly outperforms traditional approaches [1202.2395].

## 4. Extensions and Robustness

Multiple generalizations are available within the minimum-ratio estimator framework:

- **Combined estimator:** When limited labeled target data are available, the ratio estimator can be linearly combined with plug-in estimators, using weights determined by their MSEs. This blend is provably optimal in MSE for the convex mixture [1807.03929].
- **Regression-ratio estimator:** For covariate-dependent prevalences $\pi_t(z)$, the ratio estimator extends under extra conditional independence, permitting estimation via nonparametric regression fits of $E[g(X)|S=0, Z=z]$. Consistency and convergence rates are characterized under mild regularity [1807.03929].
- **Group sampling:** In two-sample experiments, minimum-ratio sampling by group or batch is handled by simulating incremental sampling, and the performance guarantees persist with negligible overshoot variation [2503.04876].

Crucially, weaker, empirically testable variants of the core assumptions (e.g., weak prior shift: $g(X) \perp S | Y$) suffice for consistency and confidence interval construction. Verification procedures based on the convex hull relationship of CDFs support practical deployment.

## 5. Optimality and Performance

For quantification, the minimax lower bound for mean squared error cannot improve upon $O(\max(n_s^{-1}, n_t^{-1}))$, as established by Bayes and Le Cam type arguments. The ratio estimator matches this rate, and hence is approximately minimax. For relative risk, the minimum-ratio estimator is unbiased and has relative mean-square error strictly bounded by $\varepsilon^2$ everywhere; for small error requirements, its efficiency (variance relative to the Cramér–Rao lower bound) converges to unity [2503.04876].

For auxiliary-based population mean estimation, the minimum-MSE estimator achieves the classical regression-type optimum, with the empirical MSE and coverage of confidence intervals demonstrating substantial relative efficiency gains over sample mean, ratio, or product estimators [1202.2395].

## 6. Algorithmic Construction and Implementation

The minimum-ratio estimator methodology is characterized by systematic algebraic derivations determining parameters—either tuning $(\alpha,\beta)$, or sequential stopping rules $(r_i, s_i)$—to meet targeted precision. For relative-risk, the construction proceeds via:

1. Solving for pilot-stage sizes ($r_1, r_2$) from curvature conditions.
2. Computing $(s_1, s_2)$ to exactly achieve the mean-square error and allocation targets.
3. Executing sampling and estimator calculations accordingly.

For quantification, the workflow is:

- Compute class means of the score function $g$ over labeled and target samples.
- Form the ratio of means estimator (possibly trimmed).
- Combine with plug-in estimators as needed or incorporate regression for prevalence as function of covariates.

The explicit forms support straightforward algorithmic implementation, including for large-scale or streaming contexts, and the underlying variance control properties remain valid under a wide range of scenarios due to their non-reliance on asymptotics or strong distributional assumptions.

## 7. Applications and Empirical Performance

Minimum-ratio estimators are applied in diverse contexts, such as:

- Quantification of prevalence in target populations under distribution shift (e.g., sentiment analysis domain adaptation) [1807.03929].
- Design-based survey estimation with auxiliary variables, especially when high correlation allows for dramatic MSE reduction [1202.2395].
- Clinical and epidemiological studies requiring strict control on the error in estimation of relative risk, regardless of underlying proportions [2503.04876].

Empirical validation demonstrates substantial improvements in MSE and coverage, robustness to misspecification of underlying distributions (given testable assumptions), and scalability to high-throughput sampling schemes.

---

The development and analysis of minimum-ratio estimators unify minimax, unbiasedness, and efficiency properties across quantification, finite-population estimation, and two-sample problems. Their construction enables both rigorous statistical guarantees and practical deployability in heterogeneous data regimes.

Source: https://www.emergentmind.com/topics/minimum-ratio-estimator