We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings.
Published online by Cambridge University Press: 27 October 2017
While applications of “big data” methods to social data have exploded, applications to social science have not. I discuss why, and I review some recent applications of big data methods to applied microeconomics. I speculate on opportunities to bring more big data methods into applied economics, and on opportunities to bring more economics to big data.
INTRODUCTION
There are n + 1 units. The first n units are the training sample. The remaining unit is the target. For each unit i in the training sample, a social scientist measures an r-dimensional response yi, a b-dimensional vector of background covariates xi, and a p-dimensional vector of policies zi. After observing the training sample and the target covariates xn+1, the social scientist chooses the target policy zn+1.
Data are called “big” when scale makes it impossible to use methods we might use easily on a smaller-scale problem. To fix ideas, think of ordinary least squares (OLS) as the canonical “small data” tool. The data may be big in the sense that we cannot load the product of the design matrix into working memory and invert it (large n). Or the data may be big in the sense that there are so many covariates that the OLS solution is ill-defined (b > n). Or the data may be big in the sense that the number of covariates grows with the sample (b growing as n grows).
The term “big data” is commonly applied to all of these cases, but especially to large n. The term “high-dimensional” is often applied to large r, b, or p.
Ongoing improvements in information and communication technology are making large-scale data increasingly available to governments and private actors, and hence to researchers. Much of this data – on Internet browsing habits, search queries, payment method use, interactions via social media, etc. – is social in nature and of obvious interest to social scientists. Yet most of the methodological advances in big data are occurring outside the social sciences.
In this article, I discuss the strengths and weaknesses of frontier big-data methods as tools for social science.
To save this book to your Kindle, first ensure [email protected] is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about saving to your Kindle.
Note you can select to save to either the @free.kindle.com or @kindle.com variations. ‘@free.kindle.com’ emails are free but can only be saved to your device when it is connected to wi-fi. ‘@kindle.com’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.
Find out more about the Kindle Personal Document Service.
To save content items to your account, please confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your account. Find out more about saving content to Dropbox.
To save content items to your account, please confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your account. Find out more about saving content to Google Drive.
