Many complexities arise when studying individuals with Autism Spectrum Disorders (ASD), including challenges in measuring key constructs as well as in estimating the effects of exposures or interventions. This panel will highlight recent advances in statistical methods that are relevant for autism research. The first talk will discuss measurement challenges and specifically how to use Item Response Theory to equate ADOS modules 1 and 2, allowing longitudinal analysis across time. The other three talks will discuss various advances in estimating causal effects: innovative randomized designs for studying adaptive interventions (with three examples of using these designs to study interventions for children with ASD), clever use of non-experimental data to estimate the effects of non-randomized exposures (applied to examining the effects of antidepressant use during pregnancy), and the use of propensity scores to equate two groups (applied to examining the associations between prenatal nutritional supplementation and ASD). Attendees will come away with a better understanding of advanced statistical methods and how new methods can help them obtain better answers to more nuanced questions of relevance to autism research.