Cookies en Macro by Mark.Usamos cookies esenciales para inicio de sesión, idioma y seguridad. Con tu permiso, también usamos cookies opcionales de preferencias, para analítica y para medición publicitaria con el píxel de Meta (Facebook). Lee nuestra política de cookies.
Extracts a small number of latent factors from many macro indicators and lets those factors drive a current-quarter read.
Follow the same equations and assumptions one section at a time or in a continuous reading.
Hover or focus a symbol defined in this model to find its exact matches. Select it to keep the highlight; Escape clears the selection.
Start with the static representation. The vector of observed variables is decomposed into a common component and an idiosyncratic component . The factors are , the loadings are , and the idiosyncratic errors are . The key identifying restriction: the common and idiosyncratic components are orthogonal.
Principal components estimation does not require specifying factor dynamics or the distribution of the errors. Stack observations into a matrix . The first principal components of are consistent estimators of the factor space as both and grow. The eigendecomposition of the sample covariance matrix yields eigenvectors with eigenvalues . The estimated loadings are , and the estimated factors are .
The static factor model. Each variable is a linear combination of r common factors plus idiosyncratic noise.
The covariance decomposition. The first eigenvalues of capture the common component.
The PCA estimate of the loadings. The normalization ensures .
The estimated factors are weighted averages of the observed variables, with weights given by the loadings.
Information criterion for selecting the number of factors. Minimize over .
Rotation indeterminacy
PCA identifies the factor space but not the individual factors. For any orthogonal matrix and give the same fit. Naming factors (calling the first one a 'demand factor') requires additional restrictions or structural identification beyond PCA.
The state is updated at observed dates. A line between dates is a display convention unless the model specifies behavior inside the interval.
Read adjacent points as adjacent sample calculations. Smoothness is a property of the estimator, not of the chart line.
Do not infer within-period behavior from a sampled estimator.
The state updates at sample dates. The visible line is only as smooth as the estimated state sequence.
Observed series can move differently from latent states because measurement noise and loadings matter.