Least squares

The result of fitting a set of data points with a quadratic function
Conic fitting a set of points using least-squares approximation

The method of least squares is a standard approach in regression analysis to the approximate solution of overdetermined systems, i.e., sets of equations in which there are more equations than unknowns. "Least squares" means that the overall solution minimizes the sum of the squares of the errors made in the results of every single equation.

The most important application is in data fitting. The best fit in the least-squares sense minimizes the sum of squared residuals, a residual being the difference between an observed value and the fitted value provided by a model. When the problem has substantial uncertainties in the independent variable (the x variable), then simple regression and least squares methods have problems; in such cases, the methodology required for fitting errors-in-variables models may be considered instead of that for least squares.

Least squares problems fall into two categories: linear or ordinary least squares and non-linear least squares, depending on whether or not the residuals are linear in all unknowns. The linear least-squares problem occurs in statistical regression analysis; it has a closed-form solution. The non-linear problem is usually solved by iterative refinement; at each iteration the system is approximated by a linear one, and thus the core calculation is similar in both cases.

Polynomial least squares describes the variance in a prediction of the dependent variable as a function of the independent variable and the deviations from the fitted curve.

When the observations come from an exponential family and mild conditions are satisfied, least-squares estimates and maximum-likelihood estimates are identical.[1] The method of least squares can also be derived as a method of moments estimator.

The following discussion is mostly presented in terms of linear functions but the use of least-squares is valid and practical for more general families of functions. Also, by iteratively applying local quadratic approximation to the likelihood (through the Fisher information), the least-squares method may be used to fit a generalized linear model.

For the topic of approximating a function by a sum of others using an objective function based on squared distances, see least squares (function approximation).

The least-squares method is usually credited to Carl Friedrich Gauss (1795),[2] but it was first published by Adrien-Marie Legendre.[3]

History

Context

The method of least squares grew out of the fields of astronomy and geodesy as scientists and mathematicians sought to provide solutions to the challenges of navigating the Earth's oceans during the Age of Exploration. The accurate description of the behavior of celestial bodies was the key to enabling ships to sail in open seas, where sailors could no longer rely on land sightings for navigation.

The method was the culmination of several advances that took place during the course of the eighteenth century:[4]

The method

The first clear and concise exposition of the method of least squares was published by Legendre in 1805.[5] The technique is described as an algebraic procedure for fitting linear equations to data and Legendre demonstrates the new method by analyzing the same data as Laplace for the shape of the earth. The value of Legendre's method of least squares was immediately recognized by leading astronomers and geodesists of the time.

In 1809 Carl Friedrich Gauss published his method of calculating the orbits of celestial bodies. In that work he claimed to have been in possession of the method of least squares since 1795. This naturally led to a priority dispute with Legendre. However, to Gauss's credit, he went beyond Legendre and succeeded in connecting the method of least squares with the principles of probability and to the normal distribution. He had managed to complete Laplace's program of specifying a mathematical form of the probability density for the observations, depending on a finite number of unknown parameters, and define a method of estimation that minimizes the error of estimation. Gauss showed that arithmetic mean is indeed the best estimate of the location parameter by changing both the probability density and the method of estimation. He then turned the problem around by asking what form the density should have and what method of estimation should be used to get the arithmetic mean as estimate of the location parameter. In this attempt, he invented the normal distribution.

An early demonstration of the strength of Gauss' Method came when it was used to predict the future location of the newly discovered asteroid Ceres. On 1 January 1801, the Italian astronomer Giuseppe Piazzi discovered Ceres and was able to track its path for 40 days before it was lost in the glare of the sun. Based on these data, astronomers desired to determine the location of Ceres after it emerged from behind the sun without solving Kepler's complicated nonlinear equations of planetary motion. The only predictions that successfully allowed Hungarian astronomer Franz Xaver von Zach to relocate Ceres were those performed by the 24-year-old Gauss using least-squares analysis.

In 1810, after reading Gauss's work, Laplace, after proving the central limit theorem, used it to give a large sample justification for the method of least square and the normal distribution. In 1822, Gauss was able to state that the least-squares approach to regression analysis is optimal in the sense that in a linear model where the errors have a mean of zero, are uncorrelated, and have equal variances, the best linear unbiased estimator of the coefficients is the least-squares estimator. This result is known as the Gauss–Markov theorem.

The idea of least-squares analysis was also independently formulated by the American Robert Adrain in 1808. In the next two centuries workers in the theory of errors and in statistics found many different ways of implementing least squares.[6]

Problem statement

The objective consists of adjusting the parameters of a model function to best fit a data set. A simple data set consists of n points (data pairs) (x_i,y_i)\!, i = 1, ..., n, where x_i\! is an independent variable and y_i\! is a dependent variable whose value is found by observation. The model function has the form f(x,\beta), where m adjustable parameters are held in the vector \boldsymbol \beta. The goal is to find the parameter values for the model which "best" fits the data. The least squares method finds its optimum when the sum, S, of squared residuals

S=\sum_{i=1}^{n}{r_i}^2

is a minimum. A residual is defined as the difference between the actual value of the dependent variable and the value predicted by the model.

r_i=y_i-f(x_i,\boldsymbol \beta).

An example of a model is that of the straight line in two dimensions. Denoting the y-intercept as \beta_0 and the slope as \beta_1, the model function is given by f(x,\boldsymbol \beta)=\beta_0+\beta_1 x. See linear least squares for a fully worked out example of this model.

A data point may consist of more than one independent variable. For example, when fitting a plane to a set of height measurements, the plane is a function of two independent variables, x and z, say. In the most general case there may be one or more independent variables and one or more dependent variables at each data point.

Limitations

This regression formulation considers only residuals in the dependent variable. There are two rather different contexts in which different implications apply:

Solving the least squares problem

The minimum of the sum of squares is found by setting the gradient to zero. Since the model contains m parameters, there are m gradient equations:

\frac{\partial S}{\partial \beta_j}=2\sum_i r_i\frac{\partial r_i}{\partial \beta_j}=0,\ j=1,\ldots,m,

and since r_i=y_i-f(x_i,\boldsymbol \beta), the gradient equations become

-2\sum_i r_i\frac{\partial f(x_i,\boldsymbol \beta)}{\partial \beta_j}=0,\ j=1,\ldots,m.

The gradient equations apply to all least squares problems. Each particular problem requires particular expressions for the model and its partial derivatives.

Linear least squares

Main article: Linear least squares

A regression model is a linear one when the model comprises a linear combination of the parameters, i.e.,

 f(x, \beta) = \sum_{j = 1}^m \beta_j \phi_j(x),

where the function \phi_{j} is a function of  x .

Letting

 X_{ij}= \frac{\partial f(x_i,\boldsymbol \beta)}{\partial \beta_j} = \phi_j(x_{i}),

we can then see that in that case the least square estimate (or estimator, in the context of a random sample),  \boldsymbol \beta is given by

 \boldsymbol{\hat\beta} =( X ^TX)^{-1}X^{T}\boldsymbol y.

For a derivation of this estimate see Linear least squares (mathematics).

Non-linear least squares

There is no closed-form solution to a non-linear least squares problem. Instead, numerical algorithms are used to find the value of the parameters \beta that minimizes the objective. Most algorithms involve choosing initial values for the parameters. Then, the parameters are refined iteratively, that is, the values are obtained by successive approximation:

{\beta_j}^{k+1}={\beta_j}^k+\Delta \beta_j,

where k is an iteration number, and the vector of increments \Delta \beta_j is called the shift vector. In some commonly used algorithms, at each iteration the model may be linearized by approximation to a first-order Taylor series expansion about  \boldsymbol \beta^k:


\begin{align}
f(x_i,\boldsymbol \beta) &= f^k(x_i,\boldsymbol \beta) +\sum_j \frac{\partial f(x_i,\boldsymbol \beta)}{\partial \beta_j} \left(\beta_j-{\beta_j}^k \right) \\
&= f^k(x_i,\boldsymbol \beta) +\sum_j J_{ij} \Delta\beta_j.
\end{align}

The Jacobian J is a function of constants, the independent variable and the parameters, so it changes from one iteration to the next. The residuals are given by

r_i=y_i- f^k(x_i,\boldsymbol \beta)- \sum_{k=1}^{m} J_{ik}\Delta\beta_k=\Delta y_i- \sum_{j=1}^{m} J_{ij}\Delta\beta_j.

To minimize the sum of squares of r_i, the gradient equation is set to zero and solved for  \Delta \beta_j:

-2\sum_{i=1}^{n}J_{ij} \left( \Delta y_i-\sum_{k=1}^{m} J_{ik}\Delta \beta_k \right) = 0,

which, on rearrangement, become m simultaneous linear equations, the normal equations:

\sum_{i=1}^{n}\sum_{k=1}^{m} J_{ij}J_{ik}\Delta \beta_k=\sum_{i=1}^{n} J_{ij}\Delta y_i \qquad (j=1,\ldots,m).

The normal equations are written in matrix notation as

\mathbf{\left(J^TJ\right)\Delta \boldsymbol \beta=J^T\Delta y}.\,

These are the defining equations of the Gauss–Newton algorithm.

Differences between linear and nonlinear least squares

These differences must be considered whenever the solution to a nonlinear least squares problem is being sought.

Least squares, regression analysis and statistics

The method of least squares is often used to generate estimators and other statistics in regression analysis.

Consider a simple example drawn from physics. A spring should obey Hooke's law which states that the extension of a spring y is proportional to the force, F, applied to it.

y = f(F,k)=kF\!

constitutes the model, where F is the independent variable. To estimate the force constant, k, a series of n measurements with different forces will produce a set of data, (F_i, y_i),\ i=1,\dots,n\!, where yi is a measured spring extension. Each experimental observation will contain some error. If we denote this error \varepsilon, we may specify an empirical model for our observations,

 y_i = kF_i + \varepsilon_i. \,

There are many methods we might use to estimate the unknown parameter k. Noting that the n equations in the m variables in our data comprise an overdetermined system with one unknown and n equations, we may choose to estimate k using least squares. The sum of squares to be minimized is

 S = \sum_{i=1}^{n} \left(y_i - kF_i\right)^2.

The least squares estimate of the force constant, k, is given by

\hat k=\frac{\sum_i F_i y_i}{\sum_i {F_i}^2}.

Here it is assumed that application of the force causes the spring to expand and, having derived the force constant by least squares fitting, the extension can be predicted from Hooke's law.

In regression analysis the researcher specifies an empirical model. For example, a very common model is the straight line model which is used to test if there is a linear relationship between dependent and independent variable. If a linear relationship is found to exist, the variables are said to be correlated. However, correlation does not prove causation, as both variables may be correlated with other, hidden, variables, or the dependent variable may "reverse" cause the independent variables, or the variables may be otherwise spuriously correlated. For example, suppose there is a correlation between deaths by drowning and the volume of ice cream sales at a particular beach. Yet, both the number of people going swimming and the volume of ice cream sales increase as the weather gets hotter, and presumably the number of deaths by drowning is correlated with the number of people going swimming. Perhaps an increase in swimmers causes both the other variables to increase.

In order to make statistical tests on the results it is necessary to make assumptions about the nature of the experimental errors. A common (but not necessary) assumption is that the errors belong to a normal distribution. The central limit theorem supports the idea that this is a good approximation in many cases.

However, if the errors are not normally distributed, a central limit theorem often nonetheless implies that the parameter estimates will be approximately normally distributed so long as the sample is reasonably large. For this reason, given the important property that the error mean is independent of the independent variables, the distribution of the error term is not an important issue in regression analysis. Specifically, it is not typically important whether the error term follows a normal distribution.

In a least squares calculation with unit weights, or in linear regression, the variance on the jth parameter, denoted \operatorname{var}(\hat{\beta}_j), is usually estimated with

\text{var}(\hat{\beta}_j)= \sigma^2\left( \left[X^TX\right]^{-1}\right)_{jj} \approx \frac{S}{n-m}\left( \left[X^TX\right]^{-1}\right)_{jj},

where the true residual variance σ2 is replaced by an estimate based on the minimised value of the sum of squares objective function S. The denominator, n  m, is the statistical degrees of freedom; see effective degrees of freedom for generalizations.

Confidence limits can be found if the probability distribution of the parameters is known, or an asymptotic approximation is made, or assumed. Likewise statistical tests on the residuals can be made if the probability distribution of the residuals is known or assumed. The probability distribution of any linear combination of the dependent variables can be derived if the probability distribution of experimental errors is known or assumed. Inference is particularly straightforward if the errors are assumed to follow a normal distribution, which implies that the parameter estimates and residuals will also be normally distributed conditional on the values of the independent variables.

Weighted least squares

A special case of generalized least squares called weighted least squares occurs when all the off-diagonal entries of Ω (the correlation matrix of the residuals) are null; the variances of the observations (along the covariance matrix diagonal) may still be unequal (heteroscedasticity).

The expressions given above are based on the implicit assumption that the errors are uncorrelated with each other and with the independent variables and have equal variance. The Gauss–Markov theorem shows that, when this is so, \hat{\boldsymbol{\beta}} is a best linear unbiased estimator (BLUE). If, however, the measurements are uncorrelated but have different uncertainties, a modified approach might be adopted. Aitken showed that when a weighted sum of squared residuals is minimized, \hat{\boldsymbol{\beta}} is the BLUE if each weight is equal to the reciprocal of the variance of the measurement

 S = \sum_{i=1}^{n} W_{ii}{r_i}^2,\qquad W_{ii}=\frac{1}{{\sigma_i}^2}

The gradient equations for this sum of squares are

-2\sum_i W_{ii}\frac{\partial f(x_i,\boldsymbol {\beta})}{\partial \beta_j} r_i=0,\qquad j=1,\ldots,n

which, in a linear least squares system give the modified normal equations,

\sum_{i=1}^{n}\sum_{k=1}^{m} X_{ij}W_{ii}X_{ik}\hat{ \beta}_k=\sum_{i=1}^{n} X_{ij}W_{ii}y_i, \qquad j=1,\ldots,m\,.

When the observational errors are uncorrelated and the weight matrix, W, is diagonal, these may be written as

\mathbf{\left(X^TWX\right)\hat {\boldsymbol {\beta}}=X^TWy}.

If the errors are correlated, the resulting estimator is the BLUE if the weight matrix is equal to the inverse of the variance-covariance matrix of the observations.

When the errors are uncorrelated, it is convenient to simplify the calculations to factor the weight matrix as \mathbf{w_{ii}}=\sqrt{\mathbf{W_{ii}}}. The normal equations can then be written in the same form as ordinary least squares:

\mathbf{\left(X'^TX'\right)\hat{\boldsymbol{\beta}}=X'^Ty'}\,

where we define the following scaled matrix and vector:

\begin{align}
\mathbf{X'} &= \text{diag}\{\mathbf{w}\} \mathbf{X},\\
\mathbf{y'} &= \text{diag}\{\mathbf{w}\} \mathbf{y}.\,\\
\end{align}

For non-linear least squares systems a similar argument shows that the normal equations should be modified as follows.

\mathbf{\left(J^TWJ\right)\boldsymbol \Delta \beta=J^TW \boldsymbol\Delta y}.\,

Note that for empirical tests, the appropriate W is not known for sure and must be estimated. For this feasible generalized least squares (FGLS) techniques may be used.

Relationship to principal components

The first principal component about the mean of a set of points can be represented by that line which most closely approaches the data points (as measured by squared distance of closest approach, i.e. perpendicular to the line). In contrast, linear least squares tries to minimize the distance in the y direction only. Thus, although the two use a similar error metric, linear least squares is a method that treats one dimension of the data preferentially, while PCA treats all dimensions equally.

Regularized versions

Tikhonov regularization

In some contexts a regularized version of the least squares solution may be preferable. Tikhonov regularization (or ridge regression) adds a constraint that \|\beta\|^2, the L2-norm of the parameter vector, is not greater than a given value. Equivalently, it may solve an unconstrained minimization of the least-squares penalty with \alpha\|\beta\|^2 added, where \alpha is a constant (this is the Lagrangian form of the constrained problem). In a Bayesian context, this is equivalent to placing a zero-mean normally distributed prior on the parameter vector.

Lasso method

An alternative regularized version of least squares is Lasso (least absolute shrinkage and selection operator), which uses the constraint that \|\beta\|_1, the L1-norm of the parameter vector, is no greater than a given value.[8][9][10] (As above, this is equivalent to an unconstrained minimization of the least-squares penalty with \alpha\|\beta\|_1 added.) In a Bayesian context, this is equivalent to placing a zero-mean Laplace prior distribution on the parameter vector.[11] The optimization problem may be solved using quadratic programming or more general convex optimization methods, as well as by specific algorithms such as the least angle regression algorithm.

One of the prime differences between Lasso and ridge regression is that in ridge regression, as the penalty is increased, all parameters are reduced while still remaining non-zero, while in Lasso, increasing the penalty will cause more and more of the parameters to be driven to zero. This is an advantage of Lasso over ridge regression, as driving parameters to zero deselects the features from the regression. Thus, Lasso automatically selects more relevant features and discards the others, whereas Ridge regression never fully discards any features. Some feature selection techniques are developed based on the LASSO including Bolasso which bootstraps samples,[12] and FeaLect which analyzes the regression coefficients corresponding to different values of \alpha to score all the features.[13]

The L1-regularized formulation is useful in some contexts due to its tendency to prefer solutions with fewer nonzero parameter values, effectively reducing the number of variables upon which the given solution is dependent.[8] For this reason, the Lasso and its variants are fundamental to the field of compressed sensing. An extension of this approach is elastic net regularization.

See also

References

  1. Charnes, A.; Frome, E. L.; Yu, P. L. (1976). "The Equivalence of Generalized Least Squares and Maximum Likelihood Estimates in the Exponential Family". Journal of the American Statistical Association 71 (353): 169–171. doi:10.1080/01621459.1976.10481508.
  2. Bretscher, Otto (1995). Linear Algebra With Applications (3rd ed.). Upper Saddle River, NJ: Prentice Hall.
  3. Stigler, Stephen M. (1981). "Gauss and the Invention of Least Squares". Ann. Stat. 9 (3): 465–474. doi:10.1214/aos/1176345451.
  4. Stigler, Stephen M. (1986). The History of Statistics: The Measurement of Uncertainty Before 1900. Cambridge, MA: Belknap Press of Harvard University Press. ISBN 0-674-40340-1.
  5. Legendre, Adrien-Marie (1805), Nouvelles méthodes pour la détermination des orbites des comètes [New Methods for the Determination of the Orbits of Comets] (in French), Paris: F. Didot
  6. Aldrich, J. (1998). "Doing Least Squares: Perspectives from Gauss and Yule". International Statistical Review 66 (1): 61–81. doi:10.1111/j.1751-5823.1998.tb00406.x.
  7. For a good introduction to error-in-variables, please see Fuller, W. A. (1987). Measurement Error Models. John Wiley & Sons. ISBN 0-471-86187-1.
  8. 1 2 Tibshirani, R. (1996). "Regression shrinkage and selection via the lasso". Journal of the Royal Statistical Society, Series B 58 (1): 267–288. JSTOR 2346178.
  9. Hastie, Trevor; Tibshirani, Robert; Friedman, Jerome H. (2009). "The Elements of Statistical Learning" (second ed.). Springer-Verlag. ISBN 978-0-387-84858-7.
  10. Bühlmann, Peter; van de Geer, Sara (2011). Statistics for High-Dimensional Data: Methods, Theory and Applications. Springer. ISBN 9783642201929.
  11. Park, Trevor; Casella, George (2008). "The Bayesian Lasso". Journal of the American Statistical Association 103 (482): 681–686. doi:10.1198/016214508000000337.
  12. Bach, Francis R (2008). "Bolasso: model consistent lasso estimation through the bootstrap". Proceedings of the 25th international conference on Machine learning: 33–40. doi:10.1145/1390156.1390161.
  13. Zare, Habil (2013). "Scoring relevancy of features based on combinatorial analysis of Lasso with application to lymphoma diagnosis". BMC Genomics 14: S14. doi:10.1186/1471-2164-14-S1-S14.

Further reading

This article is issued from Wikipedia - version of the Sunday, April 03, 2016. The text is available under the Creative Commons Attribution/Share Alike but additional terms may apply for the media files.