Approximate Bayes Factors Are Approximate E-Processes

Recent years have seen a surge of interest in 'e-values' and 'e-processes', which allow sequential inference and optional stopping while maintaining strict frequentist Type I error control. For the test of a single parameter we present families of Wald-type test statistics that are both e-processes and approximate Bayes factors (e.g., Jeffreys, 1936).