# Least Squares Estimation of Poisson Parameter

"Assume independent random variables $Y_i$~$Poisson(λx_i)$. Supposing that $x_i$ are given, fixed constants, obtain the least squares estimator of $λ$ and compute its variance."

This kind of a problem is the first one faced by me. The first things that are coming to my mind are regression equation of the form $log(Y)=a+bx$ and then trying the least square algorithm by minimizing $∑(log(y)-a-bx)^2$. From this I will be getting the least squares estimates of $a,b$ but how do I get $λ$?

As I said, this is the first time I am facing such a question. I know point estimation in terms of MLE and moment-equality. I need help and resources (i.e. more knowledge) about solving this kind of a question. Can you kindly supply me with the answer and the explanation along with maybe links to some pdf's? Google is not helping me. Thanks in advance.

(EDIT:) While I was waiting for any response to this question, I looked up deeper and found that I did not know one information. That is, the least square estimator for a parameter $θ$ is obtained by minimising $∑(Y_i-E(Y_i|x_i,θ))^2$. So I tried to work on this using my new knowledge.

I get the LSE of $λ$ as $\dfrac{∑x_iY_i}{∑(x_i)^2}$. Please tell me if this is correct.

Then the variance of this LSE is $\dfrac{λ∑(x_i)^3}{(∑(x_i)^2)^2}$.

Is this right? Thanks for your time.

• If $Y_i \sim \mbox{Poisson}(\lambda x_i)$, what is $E(Y_i | x_i)$? Could you minimize $\sum \left( Y_i - E(Y_i | x_i)\right)^2$? As far as taking the log goes, what happens if $Y_i =0$, which occurs with positive probability under a Poisson distribution? – Andrew M Oct 31 '14 at 7:22
• Dear Andrew, thank you for this information. I did not initially know that the true method is minimizing $∑(Y_i−E(Y_i|x_i))^2$. Fortunately, I searched it out myself on Google shortly before you commented. Thank you. Would you kindly verify if my answers are correct? I have edited my answer. – Landon Carter Oct 31 '14 at 7:27

### Log Likelihood Solution

The Log Likelihood Function is given by:

\begin{align*} \hat{\lambda} & = \arg \max_{\lambda} p \left( \boldsymbol{y} \mid \boldsymbol{x}, \lambda \right) \\ & = \arg \max_{\lambda} \log \prod_{i = 1}^{n} \frac{ \left( \lambda {x}_{i} \right)^{ {y}_{i} } }{ {}y_{i}! } {e}^{-\lambda {x}_{i}} \\ & = \arg \max_{\lambda} \sum_{i = 1}^{n} \log \left( \frac{ \left( \lambda {x}_{i} \right)^{ {y}_{i} } }{ {}y_{i}! } {e}^{-\lambda {x}_{i}} \right) \\ & = \arg \max_{\lambda} \log \lambda \sum_{i = 1}^{n} {y}_{i} + \sum_{i = 1}^{n} \log {x}_{i} - \lambda \sum_{i = 1}^{n} {x}_{i} -\sum_{i = 1}^{n} \log {y}_{i} ! \\ & \Rightarrow \hat{\lambda} = \frac{ \sum_{i}^{n} {y}_{i} }{ \sum_{i}^{n} {x}_{i} } \end{align*}

This indeed coincide with the classic case where ${x}_{i} = 1$ and then the MLE is the empirical average.

### Least Squares Solution

The Least Squares solution is given by:

\begin{align*} \hat{\lambda} & = \arg \min_{\lambda} \sum_{i}^{n} \left( \mathbb{E} \left[ {y}_{i} \mid {x}_{i}, \lambda \right] - {y}_{i} \right)^{2} \\ & = \arg \min_{\lambda} \sum_{i}^{n} \left( \lambda {x}_{i} - {y}_{i} \right)^{2} \\ & \Rightarrow \hat{\lambda} = \frac{ \sum_{i}^{n} {x}_{i} {y}_{i} }{ \sum_{i = 1}^{n} {x}_{i}^{2} } \end{align*}

Small simulation in MATLAB:

% Mathematics Q122153
% https://stats.stackexchange.com/questions/122153
% Least Squares Estimation of Poisson Parameter
% References:
%   1.  Poisson Distribution Wikipedia - https://en.wikipedia.org/wiki/Poisson_distribution.
%   2.  Poisson Distribution (MATLAB)- https://www.mathworks.com/help/stats/poisson-distribution.html.
%   3.  Poisson Random Numbers (MATLAB)- https://www.mathworks.com/help/stats/poissrnd.html.
% Remarks:
%   1.  sa
% TODO:
%   1.  ds
% Release Notes
% - 1.0.000     10/09/2017
%   *   First release.

%% General Parameters

run('InitScript.m');

figureIdx           = 0; %<! Continue from Question 1
figureCounterSpec   = '%04d';

generateFigures = OFF;

%% Simulation Parameters

paramLambda = 0.75;
numSamples  = 1000;

%% Algorithm Parameters

gridStartVal    = 0.01;
gridEndVal      = 2;
numGridSamples  = 2000;

%% Generate Data

vX              = randi([1, 10], [numSamples, 1]); %<! Known
vParamLambda    = paramLambda * vX;
vDataSamples    = poissrnd(vParamLambda, [numSamples, 1]);

vLambdaGrid     = linspace(gridStartVal, gridEndVal, numGridSamples);

%% Maximum Likelihood Estimator of Lambda - Brute Force

vLogLikelihood  = zeros([numGridSamples, 1]);

for ii = 1:numGridSamples
currLambda = vLambdaGrid(ii);
vLogLikelihood(ii) = log(currLambda) * sum(vDataSamples) + sum(log(vX)) - currLambda * sum(vX) - sum(log(factorial(vDataSamples)));
end

%% Maximum Likelihood Estimator of Lambda - Closed Form

paramLambdaMle = sum(vDataSamples) / sum(vX);

%% Least Squares Estimator of Lambda - Brute Force

vLsLikelihood  = zeros([numGridSamples, 1]);

for ii = 1:numGridSamples
currLambda          = vLambdaGrid(ii);
vLsLikelihood(ii)   = sum( (vDataSamples - (currLambda * vX)) .^ 2 );
end

%% Least Squares Estimator of Lambda - Closed Form

paramLambdaLs = sum(vDataSamples .* vX) / sum(vX .^ 2);

%% Analysis

hFigure     = figure('Position', figPosLarge);
hAxes       = subplot(2, 1, 1);
hLineSeries = plot(vLambdaGrid, vLogLikelihood);
set(hLineSeries, 'LineWidth', lineWidthNormal);
hLineSeries = plot([paramLambda, paramLambda], get(hAxes, 'YLim'));
set(hLineSeries, 'LineStyle', ':');
hLineSeries = plot([paramLambdaMle, paramLambdaMle], get(hAxes, 'YLim'));
set(hLineSeries, 'LineStyle', ':');
set(get(hAxes, 'Title'), 'String', {['Log Likelihood Objective Function Value vs. \lambda Hypothesis']}, ...
'FontSize', fontSizeTitle);
set(get(hAxes, 'XLabel'), 'String', '\lambda', ...
'FontSize', fontSizeAxis);
set(get(hAxes, 'YLabel'), 'String', 'Objective Function Value', ...
'FontSize', fontSizeAxis);
hLegend = ClickableLegend({['Log Likelihood Objective Function'], ['Log Likelihood Maximum'], ['Ground Truth']});
% set(hAxes, 'LooseInset', [0.07, 0.07, 0.07, 0.07]);

hAxes       = subplot(2, 1, 2);
hLineSeries = plot(vLambdaGrid, log(vLsLikelihood));
set(hLineSeries, 'LineWidth', lineWidthNormal);
hLineSeries = plot([paramLambda, paramLambda], get(hAxes, 'YLim'));
set(hLineSeries, 'LineStyle', ':');
hLineSeries = plot([paramLambdaLs, paramLambdaLs], get(hAxes, 'YLim'));
set(hLineSeries, 'LineStyle', ':');
set(get(hAxes, 'Title'), 'String', {['Log Least Squares Objective Function Value vs. \lambda Hypothesis']}, ...
'FontSize', fontSizeTitle);
set(get(hAxes, 'XLabel'), 'String', '\lambda', ...
'FontSize', fontSizeAxis);
set(get(hAxes, 'YLabel'), 'String', 'Objective Function Value', ...
'FontSize', fontSizeAxis);
hLegend = ClickableLegend({['Log Least Squares Objective Function'], ['Log Least Squares Minimum'], ['Ground Truth']});
% set(hAxes, 'LooseInset', [0.07, 0.07, 0.07, 0.07]);

if(generateFigures == ON)
saveas(hFigure,['Figure', num2str(figureIdx, figureCounterSpec), '.png']);
end

%% Restore Defaults

% set(0, 'DefaultFigureWindowStyle', 'normal');
% set(0, 'DefaultAxesLooseInset', defaultLoosInset);


The full code is available on my StackExchange Cross Validated Q122153 GitHub Repository.