Scribd
Upload
10 views1 page

Ps 1

The document outlines a problem set for a course involving various least squares methods and estimations in Matlab, with specific tasks related to sinusoidal data, weighted least squares, recursive least squares, and regularized least squares. It also includes problems on estimating parameters using maximum likelihood estimation, working with noisy measurements, and understanding linear time-invariant systems. Additionally, it encourages familiarity with Matlab techniques for optimization and statistical theorems.

Uploaded by

sauhardya dutta
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
10 views1 page

Ps 1

The document outlines a problem set for a course involving various least squares methods and estimations in Matlab, with specific tasks related to sinusoidal data, weighted least squares, recursive least squares, and regularized least squares. It also includes problems on estimating parameters using maximum likelihood estimation, working with noisy measurements, and understanding linear time-invariant systems. Additionally, it encourages familiarity with Matlab techniques for optimization and statistical theorems.

Uploaded by

sauhardya dutta
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

Problem set 1

January 28, 2025

1. Matlab LS: Consider a sinusoidal data with Gaussian noise


M M
X 2πkn X 2πkn
x(n) = a0 + ak cos( )+ bk sin( ) + w(n), n = 0, . . . , N − 1.
N N
k=1 k=1

where w(n) is WGN. Let θ = (a0 , a1 , . . . , aM , b1 , . . . , bm )T . Find the LS solution in Matlab for N = 10, 20.
Find the covariance of the LS estimator.
2. Weighted LS: Find the weighted LS solution for Problem 1 with weighting matrix W = cov−1 (w(n)).

3. Recursive LS: Let y(n) = Ax(n) + w(n), n = 1, . . . , N . Determine A using recursive LS for N = 15, 25.
4. Regularized LS: Find the regularized LS solution for Problem 1 with λ = 1 i.e., ridge regression estimator. Find
the lasso estimator and compare with ridge regularized estimator. Find the covariance of the estimator.
5. Noisy regressor: Let v(k) = v0 + nv (k), i(k) = i0 + ni (k) where nv and ni denote voltage and current noise
during measurements at time k. Let nv , ni be iid, zero mean with finite variances σv2 , σi2 . Find the value of the
resistor for different values of σv2 , σi2 and compare the resistor values.
6. Repeat problems 1,2 and 4 to determine the impulse response of a SISO discrete time LTI system from N noisy
output measurements given N sinusoidal inputs of different frequencies.

7. Consider a discrete LTI system x(k + 1) = Ax(k) + Bu(k), y(k) = Cx(k) which is observable. Find the initial
state x(0) using LS. Suppose y(k) = Cx(k) + v(k) where v(k) is WGN. Find x(0) using LS methods.
8. Nonlinear Least squares: Let y(n) = sin ωn + v(n) where v is white Gaussian. Estimate ω from N measure-
ments.
9. MLE estimators: Find the MLE solution to Problem 1.

10. MLE estimators: Consider Y ∼ N (µ, σ 2 ) and let D = {yn : n = 1 : N } be an iid sample of size N . Let
θ = (µ, σ 2 ) be the parameters to be estimated. Use MLE to estimate θ and find MLE estimators for µ and σ 2 .
Are they familiar?
11. Get familiar with QR and SVD decomposition in Matlab.

12. Get familiar with solving optimization problems in Matlab e.g., least squares problems and nonlinear optimiza-
tion problems (e.g., A/b to solve Ax = b, lsqlin, fmincon, lsqnonlin etc.).
13. Revise weak and strong law of large numbers and the central limit theorem.
14. Get familiar with the method of instrumental variables in Matlab.

You might also like