Reputation
1,082
Top tag
Next privilege 2,000 Rep.
Edit questions and answers
Badges
8 22
Newest
 Enlightened
Impact
~49k people reached

Apr
23
awarded  Enlightened
Apr
23
awarded  Nice Answer
Apr
22
comment Finding MLE by factor analyzing the correlation matrix
they have mentioned that the MLEs ($\hat{L_z}$) are obtained from the the correlation matrix $R$ by inserting $R$ in place of $[(n-1)/n] \times S$ (where $S$ is the sample covariance matrix) in the likelihood of a multivariate normal where $S_n=[(n-1)/n] \times S$ is replaced by $\hat{LL^T}+\hat{\psi}$. They say it is equivalent to setting $\hat{L_z}=\hat{V}^{-1/2}\hat{L}$.
Apr
21
comment Finding MLE by factor analyzing the correlation matrix
$\hat{V}^{-1/2}$ is the diagonal matrix with the reciprocal of the sample standard deviations (computed with divisor $\sqrt{n}$). So, the standardized variables are $Z=V^{-1/2}(X-\mu)$.
Apr
18
comment Understanding a “computationally convenient uniqueness condition” on loadings in factor analysis
Thanks I have seen some materials about SVD (on how to get it). But it would be great if you could just tell me intuitively what it does. I am actually trying to understand what was meant by 'singular value' here!
Apr
18
comment Understanding a “computationally convenient uniqueness condition” on loadings in factor analysis
No I think they don't. Can you add that too for me? Thank you for your answer.
Apr
18
accepted Understanding a “computationally convenient uniqueness condition” on loadings in factor analysis
Apr
18
comment Finding MLE by factor analyzing the correlation matrix
Actually we want to maximize the likelihood of a multivariate normal distribution where $\Sigma$ has been replaced by $\hat{LL^T}+\hat{\psi}$. Thus the MLE of the loading and specific variance parameters of a factor model are found.
Apr
16
asked Finding MLE by factor analyzing the correlation matrix
Apr
16
asked Understanding a “computationally convenient uniqueness condition” on loadings in factor analysis
Apr
14
awarded  Notable Question
Apr
10
comment Is there any general formulation procedure of probability density functions?
How to find that function? What mathematical methods are used?
Apr
9
asked Is there any general formulation procedure of probability density functions?
Mar
23
asked Meaning of weights in calculating the Myer's blended index
Mar
8
asked Problem in understanding an expectation
Mar
3
asked Some basic questions related to the moments of a probability distribution
Feb
10
awarded  Popular Question
Feb
1
accepted How to calculate probabilities when there is censoring?
Jan
26
comment How to calculate probabilities when there is censoring?
Thank you gung. But if the 2 patients got censored due to some other reason (left the study) within the study period? Should the probability estimate be 0.4 in that case?
Jan
24
comment How to calculate probabilities when there is censoring?
Actually in this clinical trial patients will either remit or will fail to remit (meaning death before remission). Some patients will get censored before remission. In the case described in the question 4 patients achieved remission, 4 people died without achieving remission and 2 people got censored before remission.