Interactive Fixed Effects Estimator for Panel Data // Implements the interactive fixed effects ('IFE') panel estimator of Bai (2009) doi:10.3982/ECTA6135 for balanced and unbalanced panels, with optional additive unit and/or time fixed effects. Provides analytical standard errors ('homoskedastic', 'HC1' heteroskedasticity-robust, cluster-robust by unit, and heteroskedasticity- and autocorrelation- consistent), together with asymptotic incidental-parameter bias correction for large panels, including a dynamic extension for predetermined (lagged-dependent) regressors following Moon and Weidner (2017) doi:10.1017/S0266466615000328. The number of factors is chosen by information criteria (Bai and Ng 2002 doi:10.1111/1468-0262.00273) or by singular value thresholding. Unbalanced panels are handled by an expectation-maximisation algorithm with nuclear-norm-regularised initialisation, with estimation, analytical inference, and bias correction following Su, Wang and Wang (2025) doi:10.2139/ssrn.5177283 and building on the matrix-completion and missing-data factor analysis of Bai and Ng (2021) doi:10.1080/01621459.2021.1967163. All computations use base R only, with no external dependencies.