We provide systematic evidence on the potential for estimating household well-being from mobile phone data. Using data from four countries - Afghanistan, Cote d'Ivoire, Malawi, and Togo - we conduct parallel, standardized machine learning experiments to assess which measures of welfare can be most accurately predicted, which types of phone data are most useful, and how much training data is required. We find that long-term poverty measures such as wealth indices (Pearson's rho = 0.20-0.59) and multidimensional poverty (rho = 0.29-0.57) can be predicted more accurately than consumption (rho = 0.04 - 0.54); transient vulnerability measures like food security and mental health are very difficult to predict. Models using calls and text message behavior are more predictive than those using metadata on mobile internet usage, mobile money transactions, and airtime top-ups. Predictive accuracy improves rapidly through the first 1,000-2,000 training observations, with continued gains beyond 4,500 observations. Model performance depends strongly on sample heterogeneity: nationally-representative samples yield 20-70 percent higher accuracy than urban-only or rural-only samples.

翻译：我们系统性地评估了利用移动电话数据估算家庭福祉的潜力。基于阿富汗、科特迪瓦、马拉维和多哥四个国家的数据，我们开展了并行化、标准化的机器学习实验，以评估哪些福利指标能被最准确地预测、哪些类型的手机数据最具效用、以及需要多少训练数据。研究发现，长期贫困指标如财富指数（皮尔逊相关系数ρ=0.20-0.59）和多维贫困指数（ρ=0.29-0.57）的预测精度高于消费指标（ρ=0.04-0.54）；而粮食安全与心理健康等短期脆弱性指标则极难预测。使用通话和短信行为数据的模型比使用移动互联网使用元数据、移动货币交易和通话时长充值数据的模型具有更强的预测能力。预测精度在前1,000-2,000个训练样本中快速提升，并在超过4,500个样本后持续改善。模型性能高度依赖样本异质性：全国代表性样本的预测准确率比纯城市或纯农村样本高出20%-70%。