Lars Peter Hansen

The challenge is to build models that are rich enough to capture the complexities of economic phenomena, yet parsimonious enough to be tractable and yield clear insights.

Uncertainty is not just about not knowing the future; it's about not knowing the true model of the world.

Robustness is a virtue in economic modeling, especially when facing model misspecification.

The generalized method of moments provides a powerful framework for estimating and testing economic models without fully specifying the data generating process.

Economists often assume too much knowledge on the part of economic agents. We need to account for their limited understanding and their desire for robustness.

Risk is when you don't know what's going to happen, but you know the probabilities. Uncertainty is when you don't even know the probabilities.

The beauty of econometrics lies in its ability to bridge economic theory and empirical observation.

We should not be afraid to admit the limitations of our models. That's where progress often begins.

The challenge for policymakers is to design policies that are robust to a wide range of plausible models of the economy.

Economic agents, like econometricians, are often concerned about model misspecification.

The concept of 'model uncertainty' is crucial for understanding how agents make decisions in complex environments.

Good economic models are not necessarily 'true' models, but rather useful approximations of reality.

The GMM framework allows us to test the implications of economic theory without making strong distributional assumptions.

Robust control theory provides a systematic way to deal with model uncertainty in policy design.

The interplay between theory and empirics is at the heart of modern econometrics.

We need to move beyond the idea of a single 'true' model and embrace the notion of a set of plausible models.

The challenge is to quantify the costs of model misspecification and design policies that minimize these costs.

Econometrics is not just about fitting curves; it's about understanding the underlying economic mechanisms.

The generalized method of moments has become a workhorse for empirical macroeconomics and finance.

A good model should be able to explain observed phenomena and make reasonable predictions, even if it's not perfectly specified.