We develop a new targeted maximum likelihood estimation method that provides improved forecasting for misspecified linear autoregressive models. The method weighs data points in the observed sample and is useful in the presence of data generating processes featuring structural breaks, complex...

According to a growing body of empirical literature, global shocks have become less important for business cycles in industrialized countries and emerging market economies since the mid-1980s. In this paper, we analyze the question of what might have caused a decoupling from the global business...

We propose an extended time-varying parameter Vector Autoregression that allows for an evolving relationship between the variances of the shocks. Using this model, we show that the relationship between the conditional variance of GDP growth and the long-term interest rate has become weaker over...

We develop a new method that detects jumps nonparametrically in financial time series and significantly outperforms the current benchmark on simulated data. We use a long short- term memory (LSTM) neural network that is trained on labelled data generated by a process that experiences both jumps...

We build a dynamic factor model with time-varying parameters and stochastic volatility and use it to decompose the variance of a large set of financial and macroeconomic variables for 22 OECD countries spanning from 1960 onwards into contributions from country-specific uncertainty,...

We develop a dynamic factor model with time-varying parameters and stochastic volatility, estimate it with several variables for a large number of countries and decompose the variance of each variable in terms of contributions from uncertainty common to all countries (global uncertainty),...