Reputation
498
Top tag
Next privilege 500 Rep.
Access review queues
Badges
3 11
Newest
 Yearling
Impact
~6k people reached

  • 0 posts edited
  • 0 helpful flags
  • 15 votes cast
1h
answered Significance of dummy variables in probit regression
1d
comment Bayesian Updating - plug in previous posterior for prior?
Just to be clear, is the new information the entire sequence or just the terms indexed by $n+1$, e.g. $a_{n+1}$ and $b_{n+1}$
1d
comment Least squares with known error in y
Measurement error in $y$ variables does not bias regression coefficients (i.e. in $x$). So, at least in theory, it should not make a difference. If that result is affected by a small sample size, I can't say for sure.
Jun
28
answered In a simple regression model estimated using OLS, the covariance between the estimated errors and regressors is zero by construction
Jun
28
comment In a simple regression model estimated using OLS, the covariance between the estimated errors and regressors is zero by construction
Yes, its true. Intercepts do not matter.
Jun
28
comment Polynomial least squares fit — restrictions on order?
The example in the link you provided is solving a simple regression with a constant term and a variable x. ($y=a+bx$) This is a regression with 2 independent variables. That's why the matrix is 2x2. (Using the formula in my previous comment, X'X is a 2 x 2 matrix). But, you want to know (or at least I think you want to know) largest order you can include in a regression of y on x, $x^2$, $x^3$, and so forth. This will require solving higher order matrices. Edit: Here's a better link. mathworld.wolfram.com/LeastSquaresFittingPolynomial.html
Jun
28
comment Polynomial least squares fit — restrictions on order?
The standard solution in a linear regression is $(X'X)^{-1}X'y$ where X is an $ N x K$ matrix with N observations and K parameters. In order to be valid the matrix $X'X$ must be invertible. This requires , by definition, that $N \geq K$.
Jun
27
revised Polynomial least squares fit — restrictions on order?
added 148 characters in body
Jun
27
answered Polynomial least squares fit — restrictions on order?
Jun
26
comment Non-linear regression fit
You can not linearize it. Not without knowing A. The closest thing you could do would be to do your method at several different values of A. And then pick the set of values with the smallest SSE.
Jun
25
comment Estimating grader bias/variance and MLE test scores given multiple graders assigned to grade each test
Yes, thats true. You might be able to use the average test score to get an additional moment. It makes sense you need to make a normalization. You need to establish some sort of baseline. For example, suppose all children received the same mark. How would you know whether all graders were biased or unbiased unless you had something to compare it to?
Jun
25
revised Estimating grader bias/variance and MLE test scores given multiple graders assigned to grade each test
added 6 characters in body
Jun
24
answered Estimating grader bias/variance and MLE test scores given multiple graders assigned to grade each test
Jun
23
comment Estimating grader bias/variance and MLE test scores given multiple graders assigned to grade each test
Are you asking what the MLE formula will be? Or are you asking how we know we will be able to estimate (i.e. identify) the parameters? Do you have a parameterization for how the biases are distributed across the graders?
Jun
20
comment Having trouble creating my Neural Network inputs
Using $ N xM $ inputs is a good idea.
Jun
20
comment Outlier detection with robust multiple regression model
Well, I don't know of any papers on the subject. The definition of an outlier is, of course, subjective. The method you described sounds fine in theory. There are other ways with dealing with extreme values. For example, you can try logging values, which will compress the distribution
Jun
20
comment Outlier detection with robust multiple regression model
Is your goal specifically to find outliers? Or is your goal just to make sure outliers are not driving your results?
Jun
18
revised Numerically solving equations with expectations
added 154 characters in body
Jun
18
answered Numerically solving equations with expectations
Jun
14
revised Derivation of Likelihood Function for Random Effects Parameters
deleted 316 characters in body