The focus of lot of recent research in causal inference has been (1) To allow for unstructured data in causal studies and (2) To better address endogeneity issues in observational data. This dissertation develops and applies non-parameteric econometeric methods to address these upcoming issues. Specifically, we (i) develop a method to address the issue of weak instruments in econometeric models, (ii) develop a causal gradient boosting based estimator to ``learn" causal effects, and (iii) apply recent empirical IO methods to study smartphone platform economy. Machine Learning Instrumental Variables for Causal Inference -- Instrumental variables (IVs) are a commonly used technique for causal inference from observational data. In practice, the variation induced by IVs can be limited, which yields imprecise or biased estimates of causal effects and renders the approach ineffective for policy decisions. We confront this challenge by formulating the problem of constructing instrumental variables from candidate exogenous data as a machine learning problem. We provide formal asymptotic theory and show root-n consistency and asymptotic efficiency of our estimators hold under very general conditions. Simulations and application to real-world data demonstrate that the algorithm is highly effective and significantly improves the performance of causal inference from observational data. Finally, we look at a recent paper by \cite{moshary2021and} that critiqued the use of political cycles as an instrument for advertising. Specifically, the authors test the strength of the first stage category-by-category for 274 product categories. The authors find that for most categories, the first-stage F-statistics are less than 10 (221 of 274 product categories) in their benchmark. We demonstrate most of the issues found by the authors were due to mis-specification issue in the first stage and can be resolved using MLIVs. Causal Gradient Boosting -- Recent advances in the literature have demonstrated that standard supervised learning algorithms are ill-suited for problems with endogenous explanatory variables. To correct for this, many variants of nonparameteric instrumental variable regression methods have been developed. In this paper, we propose an alternative algorithm called \textit{boostIV} that builds on the traditional gradient boosting algorithm and corrects for the endogeneity bias. The algorithm is very intuitive and resembles an iterative version of the standard 2SLS estimator. The proposed estimator is data driven and does not require any functional form approximation assumptions besides specifying a weak learner. We demonstrate that our estimator is consistent under mild conditions. We carry out extensive Monte Carlo simulations to demonstrate the finite sample performance of our algorithm compared to other recently developed methods. We show that boostIV is at worst on par with the existing methods and on average significantly outperforms them. Network Externalities and Platform Fees in Mobile Platforms -- The benefits consumers get from a mobile phone depending on the software apps available to use on that phone. Still, both major app platform operators -- Apple and Android charge app developers a significant platform fee to access their userbase. Recently many antitrust cases have been raised against both Apple and Android against the platform fees being charged developers to pay them commissions of up to 30% on all app store transactions. To shed some light on this ongoing legal discourse, we measure the platform fees Apple and Android should charge under market equilibrium. We also explore how the equilibrium fees vary with the strength of indirect effects and the ease of interoperating between the two platforms. To do so, we develop models of consumer demand for handsets and apps, as well as a model of developer entry decisions. We estimate the model using a unique dataset of mobile phone sales and app downloads. We find that counter to existing claims equilibrium fees is comparable to what is currently being charged by Android and Apple.