GitHub - emilylaiken/togo-targeting-replication

Source: https://github.com/emilylaiken/togo-targeting-replication

:root { --tab-size-preference: 4; } pre, code { tab-size: var(--tab-size-preference); } {"locale":"en","featureFlags":["alternate_user_config_repo","api_insights_show_missing_data_banner","attestations_filtering","attestations_sorting","client_version_header","codespaces_prebuild_region_target_update","contact_requests_implicit_opt_in","contentful_lp_enterprise","contentful_lp_footnotes","copilot_agent_task_caching","copilot_api_agentic_issue_marshal_yaml","copilot_api_draft_issue_with_mixed_trees","copilot_chat_attach_multiple_images","copilot_chat_file_redirect","copilot_chat_group_notifications","copilot_chat_model_categorization","copilot_chat_reduce_quota_checks","copilot_chat_search_bar_redirect","copilot_chat_vision_in_claude","copilot_chat_vision_skip_thread_create","copilot_custom_copilots_feature_preview","copilot_custom_copilots_images","copilot_disable_generated_commit_description","copilot_duplicate_thread","copilot_features_raycast_logo","copilot_free_to_paid_telem","copilot_ftp_hyperspace_upgrade_prompt","copilot_ftp_settings_upgrade","copilot_ftp_upgrade_to_pro_from_models","copilot_ftp_your_copilot_settings","copilot_global_overlay_v2","copilot_immersive_agent_branch_selection","copilot_immersive_draft_issue_tree","copilot_immersive_structured_model_picker","copilot_no_floating_button","copilot_read_shared_conversation","copilot_spaces_input_menu_select","copilot_spaces_permissions_tab","copilot_spaces_upsert_reload_in_background","copilot_spark_allow_empty_commit","copilot_spark_single_user_iteration","copilot_spark_use_billing_headers","copilot_spark_write_iteration_history_to_git","copilot_task_oriented_assistive_prompts","copilot_workbench_agent_seed_tool","copilot_workbench_cache","copilot_workbench_connection_reload_banner","copilot_workbench_preview_analytics","copilot_workbench_ratelimit_fallback","copilot_workbench_refresh_on_wsod","copilot_workbench_synthetic_generation","custom_copilots_capi_mode","custom_copilots_file_uploads","dashboard_public_preview","direct_to_salesforce","dotcom_chat_client_side_skills","failbot_report_error_react_apps_on_page","ghost_pilot_confidence_truncation_25","ghost_pilot_confidence_truncation_40","hpc_improve_dom_insertion_observer","hyperspace_diff_view","insert_before_patch","issue_fields_report_usage","issues_preserve_tokens_in_urls","issues_react_blur_item_picker_on_close","issues_react_bots_timeline_pagination","issues_react_prohibit_title_fallback","issues_react_remove_placeholders","lifecycle_label_name_updates","link_contact_sales_swp_marketo","marketing_pages_search_explore_provider","memex_mwl_filter_field_delimiter","migrate_toasts_to_banners_web_notifications","new_scroll_into_view","primer_react_segmented_control_tooltip","record_sso_banner_metrics","remove_child_patch","repos_insights_remove_new_url","role_assignment_list_radio","sample_network_conn_type","scheduled_reminders_updated_limits","site_homepage_contentful","site_msbuild_hide_integrations","site_msbuild_launch","site_msbuild_webgl_hero","spark_commit_on_default_branch","spark_sync_repository_after_iteration","viewscreen_sandbox","webp_support","workbench_default_sonnet4","workbench_store_readonly"],"copilotApiOverrideUrl":null}GitHub - emilylaiken/togo-targeting-replicationSkip to content{"props":{"docsUrl":"https://docs.github.com/get-started/accessibility/keyboard-shortcuts"}}

Navigation MenuToggle navigation Sign in Appearance settings{"props":{}}

• Platform

• Solutions By company size

• Resources Topics

• Open Source

• Enterprise

• PricingSearch or jump to...

Search code, repositories, users, issues, pull requests... Search ClearSearch syntax tips

Provide feedback

We read every piece of feedback, and take your input very seriously.Include my email address so I can be contacted Cancel Submit feedback

Saved searches

Use saved searches to filter your results more quicklyNameQuery

To see all available qualifiers, see our documentation. Cancel Create saved search Sign in Sign up Appearance settings{"props":{}}Resetting focusYou signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert{{ message }} emilylaiken /togo-targeting-replicationPublic

• Notifications You must be signed in to change notification settings


• Fork 10


• Star 1818 stars 10 forks BranchesTagsActivity Star Notifications You must be signed in to change notification settings


• Code


• Issues0


• Pull requests0


• Actions


• Projects0


• Security

Uh oh!

There was an error while loading. Please reload this page.

• InsightsAdditional navigation options


• Code


• Issues


• Pull requests


• Actions


• Projects


• Security


• Insights

emilylaiken/togo-targeting-replication{"props":{"initialPayload":{"allShortcutsEnabled":false,"path":"/","repo":{"id":444799217,"defaultBranch":"master","name":"togo-targeting-replication","ownerLogin":"emilylaiken","currentUserCanPush":false,"isFork":false,"isEmpty":false,"createdAt":"2022-01-05T12:41:22.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/16864441?v=4","public":true,"private":false,"isOrgOwned":false},"currentUser":null,"refInfo":{"name":"master","listCacheKey":"v0:1641386601.407285","canEdit":false,"refType":"branch","currentOid":"38f7f1e2394b9ef93d1afcccc0a5715455f826b5"},"tree":{"items":[{"name":"data","path":"data","contentType":"directory"},{"name":"outputs","path":"outputs","contentType":"directory"},{"name":"1survey.ipynb","path":"1survey.ipynb","contentType":"file"},{"name":"2satellite.ipynb","path":"2satellite.ipynb","contentType":"file"},{"name":"3ml.ipynb","path":"3ml.ipynb","contentType":"file"},{"name":"4targeting.ipynb","path":"4targeting.ipynb","contentType":"file"},{"name":"5fairness.ipynb","path":"5fairness.ipynb","contentType":"file"},{"name":"README.md","path":"README.md","contentType":"file"},{"name":"helpers.py","path":"helpers.py","contentType":"file"}],"templateDirectorySuggestionUrl":null,"readme":null,"totalCount":9,"showBranchInfobar":false},"fileTree":null,"fileTreeProcessingTime":null,"foldersToFetch":[],"treeExpanded":false,"symbolsExpanded":false,"copilotSWEAgentEnabled":false,"isOverview":true,"overview":{"banners":{"shouldRecommendReadme":false,"isPersonalRepo":false,"showUseActionBanner":false,"actionSlug":null,"actionId":null,"showProtectBranchBanner":false,"publishBannersInfo":{"dismissActionNoticePath":"/settings/dismiss-notice/publish_action_from_repo","releasePath":"/emilylaiken/togo-targeting-replication/releases/new?marketplace=true","showPublishActionBanner":false},"interactionLimitBanner":null,"showInvitationBanner":false,"inviterName":null,"actionsMigrationBannerInfo":{"releaseTags":[],"showImmutableActionsMigrationBanner":false,"initialMigrationStatus":null},"showDeployBanner":false,"detectedStack":{"framework":null,"packageManager":null}},"codeButton":{"contactPath":"/contact","isEnterprise":false,"local":{"protocolInfo":{"httpAvailable":true,"sshAvailable":null,"httpUrl":"https://github.com/emilylaiken/togo-targeting-replication.git","showCloneWarning":null,"sshUrl":null,"sshCertificatesRequired":null,"sshCertificatesAvailable":null,"ghCliUrl":"gh repo clone emilylaiken/togo-targeting-replication","defaultProtocol":"http","newSshKeyUrl":"/settings/ssh/new","setProtocolPath":"/users/set_protocol"},"platformInfo":{"cloneUrl":"https://desktop.github.com","showVisualStudioCloneButton":false,"visualStudioCloneUrl":"https://windows.github.com","showXcodeCloneButton":false,"xcodeCloneUrl":"xcode://clone?repo=https%3A%2F%2Fgithub.com%2Femilylaiken%2Ftogo-targeting-replication","zipballUrl":"/emilylaiken/togo-targeting-replication/archive/refs/heads/master.zip"}},"newCodespacePath":"/codespaces/new?hide_repo_select=true\u0026repo=444799217"},"popovers":{"rename":null,"renamedParentRepo":null},"commitCount":"22","overviewFiles":[{"displayName":"README.md","repoName":"togo-targeting-replication","refName":"master","path":"README.md","preferredFileType":"readme","tabName":"README","richText":"\u003carticle class=\"markdown-body entry-content container-lg\" itemprop=\"text\"\u003e\u003cdiv class=\"markdown-heading\" dir=\"auto\"\u003e\u003ch1 tabindex=\"-1\" class=\"heading-element\" dir=\"auto\"\u003eTogo Targeting Replication Code\u003c/h1\u003e\u003ca id=\"user-content-togo-targeting-replication-code\" class=\"anchor\" aria-label=\"Permalink: Togo Targeting Replication Code\" href=\"#togo-targeting-replication-code\"\u003e\u003csvg class=\"octicon octicon-link\" viewBox=\"0 0 16 16\" version=\"1.1\" width=\"16\" height=\"16\" aria-hidden=\"true\"\u003e\u003cpath d=\"m7.775 3.275 1.25-1.25a3.5 3.5 0 1 1 4.95 4.95l-2.5 2.5a3.5 3.5 0 0 1-4.95 0 .751.751 0 0 1 .018-1.042.751.751 0 0 1 1.042-.018 1.998 1.998 0 0 0 2.83 0l2.5-2.5a2.002 2.002 0 0 0-2.83-2.83l-1.25 1.25a.751.751 0 0 1-1.042-.018.751.751 0 0 1-.018-1.042Zm-4.69 9.64a1.998 1.998 0 0 0 2.83 0l1.25-1.25a.751.751 0 0 1 1.042.018.751.751 0 0 1 .018 1.042l-1.25 1.25a3.5 3.5 0 1 1-4.95-4.95l2.5-2.5a3.5 3.5 0 0 1 4.95 0 .751.751 0 0 1-.018 1.042.751.751 0 0 1-1.042.018 1.998 1.998 0 0 0-2.83 0l-2.5 2.5a1.998 1.998 0 0 0 0 2.83Z\"\u003e\u003c/path\u003e\u003c/svg\u003e\u003c/a\u003e\u003c/div\u003e\n\u003cdiv class=\"markdown-heading\" dir=\"auto\"\u003e\u003ch2 tabindex=\"-1\" class=\"heading-element\" dir=\"auto\"\u003eIntroduction\u003c/h2\u003e\u003ca id=\"user-content-introduction\" class=\"anchor\" aria-label=\"Permalink: Introduction\" href=\"#introduction\"\u003e\u003csvg class=\"octicon octicon-link\" viewBox=\"0 0 16 16\" version=\"1.1\" width=\"16\" height=\"16\" aria-hidden=\"true\"\u003e\u003cpath d=\"m7.775 3.275 1.25-1.25a3.5 3.5 0 1 1 4.95 4.95l-2.5 2.5a3.5 3.5 0 0 1-4.95 0 .751.751 0 0 1 .018-1.042.751.751 0 0 1 1.042-.018 1.998 1.998 0 0 0 2.83 0l2.5-2.5a2.002 2.002 0 0 0-2.83-2.83l-1.25 1.25a.751.751 0 0 1-1.042-.018.751.751 0 0 1-.018-1.042Zm-4.69 9.64a1.998 1.998 0 0 0 2.83 0l1.25-1.25a.751.751 0 0 1 1.042.018.751.751 0 0 1 .018 1.042l-1.25 1.25a3.5 3.5 0 1 1-4.95-4.95l2.5-2.5a3.5 3.5 0 0 1 4.95 0 .751.751 0 0 1-.018 1.042.751.751 0 0 1-1.042.018 1.998 1.998 0 0 0-2.83 0l-2.5 2.5a1.998 1.998 0 0 0 0 2.83Z\"\u003e\u003c/path\u003e\u003c/svg\u003e\u003c/a\u003e\u003c/div\u003e\n\u003cp dir=\"auto\"\u003eThis repository contains replication code for the paper \"Machine Learning and Mobile Phone Data Can Improve Targeting of Humanitarian Assistance\" by Emily Aiken, Suzanne Bellue, Dean Karlan, Chris Udry, and Joshua Blumenstock. This readme provides information about the code structure, including where replication code for each figure and table in the paper is located, and information about the data available.\u003c/p\u003e\n\u003cdiv class=\"markdown-heading\" dir=\"auto\"\u003e\u003ch2 tabindex=\"-1\" class=\"heading-element\" dir=\"auto\"\u003eCode Structure\u003c/h2\u003e\u003ca id=\"user-content-code-structure\" class=\"anchor\" aria-label=\"Permalink: Code Structure\" href=\"#code-structure\"\u003e\u003csvg class=\"octicon octicon-link\" viewBox=\"0 0 16 16\" version=\"1.1\" width=\"16\" height=\"16\" aria-hidden=\"true\"\u003e\u003cpath d=\"m7.775 3.275 1.25-1.25a3.5 3.5 0 1 1 4.95 4.95l-2.5 2.5a3.5 3.5 0 0 1-4.95 0 .751.751 0 0 1 .018-1.042.751.751 0 0 1 1.042-.018 1.998 1.998 0 0 0 2.83 0l2.5-2.5a2.002 2.002 0 0 0-2.83-2.83l-1.25 1.25a.751.751 0 0 1-1.042-.018.751.751 0 0 1-.018-1.042Zm-4.69 9.64a1.998 1.998 0 0 0 2.83 0l1.25-1.25a.751.751 0 0 1 1.042.018.751.751 0 0 1 .018 1.042l-1.25 1.25a3.5 3.5 0 1 1-4.95-4.95l2.5-2.5a3.5 3.5 0 0 1 4.95 0 .751.751 0 0 1-.018 1.042.751.751 0 0 1-1.042.018 1.998 1.998 0 0 0-2.83 0l-2.5 2.5a1.998 1.998 0 0 0 0 2.83Z\"\u003e\u003c/path\u003e\u003c/svg\u003e\u003c/a\u003e\u003c/div\u003e\n\u003cp dir=\"auto\"\u003eAll code is written in Jupyter Notebooks and raw Python, using Python version 3.6. Replication code for all figures and tables that are generated with code is provided. It is assumed that the reader already has access to featurized mobile phone data matched to survey observations (as shown in the synthetic data provided, see section on \"data structure\" for more information). The scripts are organized into five notebooks as follows. All notebooks call helper functions from the file \u003ccode\u003ehelpers.py\u003c/code\u003e.\u003c/p\u003e\n\u003cdiv class=\"markdown-heading\" dir=\"auto\"\u003e\u003ch4 tabindex=\"-1\" class=\"heading-element\" dir=\"auto\"\u003e\u003ccode\u003e1survey.ipynb\u003c/code\u003e\u003c/h4\u003e\u003ca id=\"user-content-1surveyipynb\" class=\"anchor\" aria-label=\"Permalink: 1survey.ipynb\" href=\"#1surveyipynb\"\u003e\u003csvg class=\"octicon octicon-link\" viewBox=\"0 0 16 16\" version=\"1.1\" width=\"16\" height=\"16\" aria-hidden=\"true\"\u003e\u003cpath d=\"m7.775 3.275 1.25-1.25a3.5 3.5 0 1 1 4.95 4.95l-2.5 2.5a3.5 3.5 0 0 1-4.95 0 .751.751 0 0 1 .018-1.042.751.751 0 0 1 1.042-.018 1.998 1.998 0 0 0 2.83 0l2.5-2.5a2.002 2.002 0 0 0-2.83-2.83l-1.25 1.25a.751.751 0 0 1-1.042-.018.751.751 0 0 1-.018-1.042Zm-4.69 9.64a1.998 1.998 0 0 0 2.83 0l1.25-1.25a.751.751 0 0 1 1.042.018.751.751 0 0 1 .018 1.042l-1.25 1.25a3.5 3.5 0 1 1-4.95-4.95l2.5-2.5a3.5 3.5 0 0 1 4.95 0 .751.751 0 0 1-.018 1.042.751.751 0 0 1-1.042.018 1.998 1.998 0 0 0-2.83 0l-2.5 2.5a1.998 1.998 0 0 0 0 2.83Z\"\u003e\u003c/path\u003e\u003c/svg\u003e\u003c/a\u003e\u003c/div\u003e\n\u003cp dir=\"auto\"\u003eCode for all figures and tables that are generated from mainly from survey data, including the calculation of the PMT and asset index, statistics on phone ownership, summary statistics from surveys, and information on weighting and response weights. Replication code for the following tables and figures is included in this notebook:\u003c/p\u003e\n\u003cul dir=\"auto\"\u003e\n\u003cli\u003eSupplementary Figure 1\u003c/li\u003e\n\u003cli\u003eSupplementary Figure 2\u003c/li\u003e\n\u003cli\u003eSupplementary Figure 3\u003c/li\u003e\n\u003cli\u003eSupplementary Figure 4\u003c/li\u003e\n\u003cli\u003eSupplementary Figure 10\u003c/li\u003e\n\u003cli\u003eSupplementary Figure 11\u003c/li\u003e\n\u003cli\u003eSupplementary Table 2\u003c/li\u003e\n\u003cli\u003eSupplementary Table 3\u003c/li\u003e\n\u003cli\u003eSupplementary Table 4\u003c/li\u003e\n\u003cli\u003eSupplementary Table 5\u003c/li\u003e\n\u003cli\u003eSupplementary Table 6\u003c/li\u003e\n\u003cli\u003eSupplementary Table 10\u003c/li\u003e\n\u003c/ul\u003e\n\u003cdiv class=\"markdown-heading\" dir=\"auto\"\u003e\u003ch4 tabindex=\"-1\" class=\"heading-element\" dir=\"auto\"\u003e\u003ccode\u003e2satellite.ipynb\u003c/code\u003e\u003c/h4\u003e\u003ca id=\"user-content-2satelliteipynb\" class=\"anchor\" aria-label=\"Permalink: 2satellite.ipynb\" href=\"#2satelliteipynb\"\u003e\u003csvg class=\"octicon octicon-link\" viewBox=\"0 0 16 16\" version=\"1.1\" width=\"16\" height=\"16\" aria-hidden=\"true\"\u003e\u003cpath d=\"m7.775 3.275 1.25-1.25a3.5 3.5 0 1 1 4.95 4.95l-2.5 2.5a3.5 3.5 0 0 1-4.95 0 .751.751 0 0 1 .018-1.042.751.751 0 0 1 1.042-.018 1.998 1.998 0 0 0 2.83 0l2.5-2.5a2.002 2.002 0 0 0-2.83-2.83l-1.25 1.25a.751.751 0 0 1-1.042-.018.751.751 0 0 1-.018-1.042Zm-4.69 9.64a1.998 1.998 0 0 0 2.83 0l1.25-1.25a.751.751 0 0 1 1.042.018.751.751 0 0 1 .018 1.042l-1.25 1.25a3.5 3.5 0 1 1-4.95-4.95l2.5-2.5a3.5 3.5 0 0 1 4.95 0 .751.751 0 0 1-.018 1.042.751.751 0 0 1-1.042.018 1.998 1.998 0 0 0-2.83 0l-2.5 2.5a1.998 1.998 0 0 0 0 2.83Z\"\u003e\u003c/path\u003e\u003c/svg\u003e\u003c/a\u003e\u003c/div\u003e\n\u003cp dir=\"auto\"\u003eCode for aggregation of satellite-based wealth estimates and satellite-based population density estimates for high-resolution poverty mapping. Replication code for the following tables and figures is included in this notebook:\u003c/p\u003e\n\u003cul dir=\"auto\"\u003e\n\u003cli\u003eExtended Data Figure 1 Panel A\u003c/li\u003e\n\u003cli\u003eSupplementary Figure 5\u003c/li\u003e\n\u003c/ul\u003e\n\u003cdiv class=\"markdown-heading\" dir=\"auto\"\u003e\u003ch4 tabindex=\"-1\" class=\"heading-element\" dir=\"auto\"\u003e\u003ccode\u003e3ml.ipynb\u003c/code\u003e\u003c/h4\u003e\u003ca id=\"user-content-3mlipynb\" class=\"anchor\" aria-label=\"Permalink: 3ml.ipynb\" href=\"#3mlipynb\"\u003e\u003csvg class=\"octicon octicon-link\" viewBox=\"0 0 16 16\" version=\"1.1\" width=\"16\" height=\"16\" aria-hidden=\"true\"\u003e\u003cpath d=\"m7.775 3.275 1.25-1.25a3.5 3.5 0 1 1 4.95 4.95l-2.5 2.5a3.5 3.5 0 0 1-4.95 0 .751.751 0 0 1 .018-1.042.751.751 0 0 1 1.042-.018 1.998 1.998 0 0 0 2.83 0l2.5-2.5a2.002 2.002 0 0 0-2.83-2.83l-1.25 1.25a.751.751 0 0 1-1.042-.018.751.751 0 0 1-.018-1.042Zm-4.69 9.64a1.998 1.998 0 0 0 2.83 0l1.25-1.25a.751.751 0 0 1 1.042.018.751.751 0 0 1 .018 1.042l-1.25 1.25a3.5 3.5 0 1 1-4.95-4.95l2.5-2.5a3.5 3.5 0 0 1 4.95 0 .751.751 0 0 1-.018 1.042.751.751 0 0 1-1.042.018 1.998 1.998 0 0 0-2.83 0l-2.5 2.5a1.998 1.998 0 0 0 0 2.83Z\"\u003e\u003c/path\u003e\u003c/svg\u003e\u003c/a\u003e\u003c/div\u003e\n\u003cp dir=\"auto\"\u003eCode for machine learning from featurized mobile phone data, including matching survey observations to mobile phone records, cross-validation for parameter selection and out-of-sample evaluation, and evaluating feature importances. Replication code for the following tables and figures is included in this notebook:\u003c/p\u003e\n\u003cul dir=\"auto\"\u003e\n\u003cli\u003eSupplementary Figure 7\u003c/li\u003e\n\u003cli\u003eSupplementary Figure 8\u003c/li\u003e\n\u003cli\u003eExtended Data Table 3\u003c/li\u003e\n\u003cli\u003eExtended Data Table 6\u003c/li\u003e\n\u003cli\u003eSupplementary Table 13\u003c/li\u003e\n\u003c/ul\u003e\n\u003cdiv class=\"markdown-heading\" dir=\"auto\"\u003e\u003ch4 tabindex=\"-1\" class=\"heading-element\" dir=\"auto\"\u003e\u003ccode\u003e4targeting.ipynb\u003c/code\u003e\u003c/h4\u003e\u003ca id=\"user-content-4targetingipynb\" class=\"anchor\" aria-label=\"Permalink: 4targeting.ipynb\" href=\"#4targetingipynb\"\u003e\u003csvg class=\"octicon octicon-link\" viewBox=\"0 0 16 16\" version=\"1.1\" width=\"16\" height=\"16\" aria-hidden=\"true\"\u003e\u003cpath d=\"m7.775 3.275 1.25-1.25a3.5 3.5 0 1 1 4.95 4.95l-2.5 2.5a3.5 3.5 0 0 1-4.95 0 .751.751 0 0 1 .018-1.042.751.751 0 0 1 1.042-.018 1.998 1.998 0 0 0 2.83 0l2.5-2.5a2.002 2.002 0 0 0-2.83-2.83l-1.25 1.25a.751.751 0 0 1-1.042-.018.751.751 0 0 1-.018-1.042Zm-4.69 9.64a1.998 1.998 0 0 0 2.83 0l1.25-1.25a.751.751 0 0 1 1.042.018.751.751 0 0 1 .018 1.042l-1.25 1.25a3.5 3.5 0 1 1-4.95-4.95l2.5-2.5a3.5 3.5 0 0 1 4.95 0 .751.751 0 0 1-.018 1.042.751.751 0 0 1-1.042.018 1.998 1.998 0 0 0-2.83 0l-2.5 2.5a1.998 1.998 0 0 0 0 2.83Z\"\u003e\u003c/path\u003e\u003c/svg\u003e\u003c/a\u003e\u003c/div\u003e\n\u003cp dir=\"auto\"\u003eCode for targeting simulations, including producing targeting tables of all kinds, ROC curves and precision vs. recall curves, and analysis of social welfare. Replication code for the following tables and figures is included in this notebook:\u003c/p\u003e\n\u003cul dir=\"auto\"\u003e\n\u003cli\u003eFigure 1\u003c/li\u003e\n\u003cli\u003eFigure 2\u003c/li\u003e\n\u003cli\u003eExtended Data Figure 1 Panel B\u003c/li\u003e\n\u003cli\u003eExtended Data Figure 4\u003c/li\u003e\n\u003cli\u003eSupplementary Figure 9\u003c/li\u003e\n\u003cli\u003eTable 1\u003c/li\u003e\n\u003cli\u003eExtended Data Table 1\u003c/li\u003e\n\u003cli\u003eExtended Data Table 2\u003c/li\u003e\n\u003cli\u003eExtended Data Table 4\u003c/li\u003e\n\u003cli\u003eSupplementary Table 1\u003c/li\u003e\n\u003cli\u003eSupplementary Table 7\u003c/li\u003e\n\u003cli\u003eSupplementary Table 8\u003c/li\u003e\n\u003cli\u003eSupplementary Table 9\u003c/li\u003e\n\u003c/ul\u003e\n\u003cdiv class=\"markdown-heading\" dir=\"auto\"\u003e\u003ch4 tabindex=\"-1\" class=\"heading-element\" dir=\"auto\"\u003e\u003ccode\u003e5fairness.ipynb\u003c/code\u003e\u003c/h4\u003e\u003ca id=\"user-content-5fairnessipynb\" class=\"anchor\" aria-label=\"Permalink: 5fairness.ipynb\" href=\"#5fairnessipynb\"\u003e\u003csvg class=\"octicon octicon-link\" viewBox=\"0 0 16 16\" version=\"1.1\" width=\"16\" height=\"16\" aria-hidden=\"true\"\u003e\u003cpath d=\"m7.775 3.275 1.25-1.25a3.5 3.5 0 1 1 4.95 4.95l-2.5 2.5a3.5 3.5 0 0 1-4.95 0 .751.751 0 0 1 .018-1.042.751.751 0 0 1 1.042-.018 1.998 1.998 0 0 0 2.83 0l2.5-2.5a2.002 2.002 0 0 0-2.83-2.83l-1.25 1.25a.751.751 0 0 1-1.042-.018.751.751 0 0 1-.018-1.042Zm-4.69 9.64a1.998 1.998 0 0 0 2.83 0l1.25-1.25a.751.751 0 0 1 1.042.018.751.751 0 0 1 .018 1.042l-1.25 1.25a3.5 3.5 0 1 1-4.95-4.95l2.5-2.5a3.5 3.5 0 0 1 4.95 0 .751.751 0 0 1-.018 1.042.751.751 0 0 1-1.042.018 1.998 1.998 0 0 0-2.83 0l-2.5 2.5a1.998 1.998 0 0 0 0 2.83Z\"\u003e\u003c/path\u003e\u003c/svg\u003e\u003c/a\u003e\u003c/div\u003e\n\u003cp dir=\"auto\"\u003eCode for fairness audits of targeting algorithms across potentially sensitive characteristics. Replication code for the following tables and figures is included in this notebook:\u003c/p\u003e\n\u003cul dir=\"auto\"\u003e\n\u003cli\u003eFigure 3\u003c/li\u003e\n\u003cli\u003eExtended Data Figure 2\u003c/li\u003e\n\u003cli\u003eExtended Data Figure 3\u003c/li\u003e\n\u003c/ul\u003e\n\u003cdiv class=\"markdown-heading\" dir=\"auto\"\u003e\u003ch2 tabindex=\"-1\" class=\"heading-element\" dir=\"auto\"\u003eData Structure\u003c/h2\u003e\u003ca id=\"user-content-data-structure\" class=\"anchor\" aria-label=\"Permalink: Data Structure\" href=\"#data-structure\"\u003e\u003csvg class=\"octicon octicon-link\" viewBox=\"0 0 16 16\" version=\"1.1\" width=\"16\" height=\"16\" aria-hidden=\"true\"\u003e\u003cpath d=\"m7.775 3.275 1.25-1.25a3.5 3.5 0 1 1 4.95 4.95l-2.5 2.5a3.5 3.5 0 0 1-4.95 0 .751.751 0 0 1 .018-1.042.751.751 0 0 1 1.042-.018 1.998 1.998 0 0 0 2.83 0l2.5-2.5a2.002 2.002 0 0 0-2.83-2.83l-1.25 1.25a.751.751 0 0 1-1.042-.018.751.751 0 0 1-.018-1.042Zm-4.69 9.64a1.998 1.998 0 0 0 2.83 0l1.25-1.25a.751.751 0 0 1 1.042.018.751.751 0 0 1 .018 1.042l-1.25 1.25a3.5 3.5 0 1 1-4.95-4.95l2.5-2.5a3.5 3.5 0 0 1 4.95 0 .751.751 0 0 1-.018 1.042.751.751 0 0 1-1.042.018 1.998 1.998 0 0 0-2.83 0l-2.5 2.5a1.998 1.998 0 0 0 0 2.83Z\"\u003e\u003c/path\u003e\u003c/svg\u003e\u003c/a\u003e\u003c/div\u003e\n\u003cp dir=\"auto\"\u003eData files that are publically available are included in this repo; for datasets that are not publically available we have included synthetic (randomly generated) data in the same format and with the same schema. Synthetic data are produced with the notebook `data/generate_synthetic_data.ipynb'. The data files are as follows:\u003c/p\u003e\n\u003cul dir=\"auto\"\u003e\n\u003cli\u003e\u003ccode\u003edata/survey2018.csv\u003c/code\u003e and \u003ccode\u003edata/survey2020.csv\u003c/code\u003e: Synthetic data with the schema of the 2018 and 2020 survey datasets, respectively. The 2018 survey is a household survey dataset; the 2020 survey is at the individual level.\u003c/li\u003e\n\u003cli\u003e\u003ccode\u003esurvey_indiv2018.csv\u003c/code\u003e: Synthetic individual-level survey data associated with the households from the 2018 survey dataset; used only to calculate statistics on individual-level phone ownership.\u003c/li\u003e\n\u003cli\u003e\u003ccode\u003edata/features2018.csv\u003c/code\u003e and \u003ccode\u003edata/features2020.csv\u003c/code\u003e: Synthetic data representing a set of featurized mobile phone data. In this file features are randomly generated; in reality they are calculated from raw mobile phone records using open source library bandicoot. Features are provided for a subset of the phone numbers (observations) in each of the 2018 and 2020 synthetic survey datasets.\u003c/li\u003e\n\u003cli\u003e\u003ccode\u003edata/single_feature2018.csv\u003c/code\u003e: Synthetic data representing the \"single mobile phone feature\" used as a counterfactual targeting method in the paper. The single mobile phone feature is provided for the same set of phone numbers that are associated with full mobile phone featurization.\u003c/li\u003e\n\u003cli\u003e\u003ccode\u003edata/inferred_home_location2018.csv\u003c/code\u003e: Synthetic data representing inferred home locations from mobile phone data. Here home locations are chosen at random; in reality they are inferred from raw mobile phone metadata. Home locations are provided for the same subset of survey observations that are associated with full mobile phone features; they are provided at the prefecture and canton level.\u003c/li\u003e\n\u003cli\u003e\u003ccode\u003edata/shapefiles\u003c/code\u003e: Shapefiles used for poverty mapping. Shapefiles are publicly available from the Humantiarian Data Exchange (\u003ca href=\"https://data.humdata.org/dataset/togo-cod-ab\" rel=\"nofollow\"\u003ehttps://data.humdata.org/dataset/togo-cod-ab\u003c/a\u003e).\u003c/li\u003e\n\u003cli\u003e\n\u003cul dir=\"auto\"\u003e\n\u003cli\u003e\u003ccode\u003edata/shapefiles/regions.geojson\u003c/code\u003e: Shapefile for regions (admin-1 level)\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cul dir=\"auto\"\u003e\n\u003cli\u003e\u003ccode\u003edata/shapefiles/prefectures.geojson\u003c/code\u003e: Shapefile for prefectures (admin-2 level)\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cul dir=\"auto\"\u003e\n\u003cli\u003e\u003ccode\u003edata/shapefiles/cantons.geojson\u003c/code\u003e: Shapefile for cantons (admin-3 level)\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\u003ccode\u003edata/satellite\u003c/code\u003e: Data for satellite-based poverty mapping.\u003c/li\u003e\n\u003cli\u003e\n\u003cul dir=\"auto\"\u003e\n\u003cli\u003e\u003ccode\u003edata/satellite/wealth.csv\u003c/code\u003e: Tile-level relative wealth estimates from satellite imagery publicly available on the Humanitarian Data Exchange (\u003ca href=\"https://data.humdata.org/dataset/relative-wealth-index\" rel=\"nofollow\"\u003ehttps://data.humdata.org/dataset/relative-wealth-index\u003c/a\u003e)\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cul dir=\"auto\"\u003e\n\u003cli\u003e\u003ccode\u003edata/satellite/pop.csv\u003c/code\u003e: Tile-level population density estimates from satellite imagery publicly available on the Humanitarian Data Exchange (\u003ca href=\"https://data.humdata.org/dataset/highresolutionpopulationdensitymaps-tgo\" rel=\"nofollow\"\u003ehttps://data.humdata.org/dataset/highresolutionpopulationdensitymaps-tgo\u003c/a\u003e)\u003c/li\u003e\n\u003c/ul\u003e\n\u003c/li\u003e\n\u003c/ul\u003e\n\u003cdiv class=\"markdown-heading\" dir=\"auto\"\u003e\u003ch2 tabindex=\"-1\" class=\"heading-element\" dir=\"auto\"\u003eRunning the Code\u003c/h2\u003e\u003ca id=\"user-content-running-the-code\" class=\"anchor\" aria-label=\"Permalink: Running the Code\" href=\"#running-the-code\"\u003e\u003csvg class=\"octicon octicon-link\" viewBox=\"0 0 16 16\" version=\"1.1\" width=\"16\" height=\"16\" aria-hidden=\"true\"\u003e\u003cpath d=\"m7.775 3.275 1.25-1.25a3.5 3.5 0 1 1 4.95 4.95l-2.5 2.5a3.5 3.5 0 0 1-4.95 0 .751.751 0 0 1 .018-1.042.751.751 0 0 1 1.042-.018 1.998 1.998 0 0 0 2.83 0l2.5-2.5a2.002 2.002 0 0 0-2.83-2.83l-1.25 1.25a.751.751 0 0 1-1.042-.018.751.751 0 0 1-.018-1.042Zm-4.69 9.64a1.998 1.998 0 0 0 2.83 0l1.25-1.25a.751.751 0 0 1 1.042.018.751.751 0 0 1 .018 1.042l-1.25 1.25a3.5 3.5 0 1 1-4.95-4.95l2.5-2.5a3.5 3.5 0 0 1 4.95 0 .751.751 0 0 1-.018 1.042.751.751 0 0 1-1.042.018 1.998 1.998 0 0 0-2.83 0l-2.5 2.5a1.998 1.998 0 0 0 0 2.83Z\"\u003e\u003c/path\u003e\u003c/svg\u003e\u003c/a\u003e\u003c/div\u003e\n\u003cp dir=\"auto\"\u003eAfter installing the necessary packages, the code will run out of the box (using the data files located in the \u003ccode\u003edata\u003c/code\u003e subfolder). Outputs will be written to the \u003ccode\u003eoutputs\u003c/code\u003e subfolder, divided into output folders for \u003ccode\u003eml\u003c/code\u003e, \u003ccode\u003esurvey\u003c/code\u003e, and \u003ccode\u003etargeting\u003c/code\u003e. Many of the figures and tables are not written to files but rather displayed only in the Jupyter notebooks.\u003c/p\u003e\n\u003c/article\u003e","loaded":true,"timedOut":false,"errorMessage":null,"headerInfo":{"toc":[{"level":1,"text":"Togo Targeting Replication Code","anchor":"togo-targeting-replication-code","htmlText":"Togo Targeting Replication Code"},{"level":2,"text":"Introduction","anchor":"introduction","htmlText":"Introduction"},{"level":2,"text":"Code Structure","anchor":"code-structure","htmlText":"Code Structure"},{"level":4,"text":"1survey.ipynb","anchor":"1surveyipynb","htmlText":"1survey.ipynb"},{"level":4,"text":"2satellite.ipynb","anchor":"2satelliteipynb","htmlText":"2satellite.ipynb"},{"level":4,"text":"3ml.ipynb","anchor":"3mlipynb","htmlText":"3ml.ipynb"},{"level":4,"text":"4targeting.ipynb","anchor":"4targetingipynb","htmlText":"4targeting.ipynb"},{"level":4,"text":"5fairness.ipynb","anchor":"5fairnessipynb","htmlText":"5fairness.ipynb"},{"level":2,"text":"Data Structure","anchor":"data-structure","htmlText":"Data Structure"},{"level":2,"text":"Running the Code","anchor":"running-the-code","htmlText":"Running the Code"}],"siteNavLoginPath":"/login?return_to=https%3A%2F%2Fgithub.com%2Femilylaiken%2Ftogo-targeting-replication"}}],"overviewFilesProcessingTime":0,"copilotSWEAgentEnabled":false}},"appPayload":{"helpUrl":"https://docs.github.com","findFileWorkerPath":"/assets-cdn/worker/find-file-worker-9bd411a8e273.js","findInFileWorkerPath":"/assets-cdn/worker/find-in-file-worker-410714137fc9.js","githubDevUrl":null,"enabled_features":{"copilot_workspace":null,"code_nav_ui_events":false,"react_blob_overlay":false,"accessible_code_button":true}}}}.hUCRAk{display:-webkit-box;display:-webkit-flex;display:-ms-flexbox;display:flex;-webkit-flex-direction:column;-ms-flex-direction:column;flex-direction:column;-webkit-align-items:center;-webkit-box-align:center;-ms-flex-align:center;align-items:center;}/*!sc*/ data-styled.g1[id="Box-sc-g0xbh4-0"]{content:"hUCRAk,"}/*!sc*/ .hWlpPn{position:relative;display:inline-block;}/*!sc*/ .hWlpPn::after{position:absolute;z-index:1000000;display:none;padding:0.5em 0.75em;font:normal normal 11px/1.5 -apple-system,BlinkMacSystemFont,"Segoe UI","Noto Sans",Helvetica,Arial,sans-serif,"Apple Color Emoji","Segoe UI Emoji";-webkit-font-smoothing:subpixel-antialiased;color:var(--tooltip-fgColor,var(--fgColor-onEmphasis,var(--color-fg-on-emphasis,#ffffff)));text-align:center;-webkit-text-decoration:none;text-decoration:none;text-shadow:none;text-transform:none;-webkit-letter-spacing:normal;-moz-letter-spacing:normal;-ms-letter-spacing:normal;letter-spacing:normal;word-wrap:break-word;white-space:pre;pointer-events:none;content:attr(aria-label);background:var(--tooltip-bgColor,var(--bgColor-emphasis,var(--color-neutral-emphasis-plus,#24292f)));border-radius:6px;opacity:0;}/*!sc*/ @-webkit-keyframes tooltip-appear{from{opacity:0;}to{opacity:1;}}/*!sc*/ @keyframes tooltip-appear{from{opacity:0;}to{opacity:1;}}/*!sc*/ .hWlpPn:hover::after,.hWlpPn:active::after,.hWlpPn:focus::after,.hWlpPn:focus-within::after{display:inline-block;-webkit-text-decoration:none;text-decoration:none;-webkit-animation-name:tooltip-appear;animation-name:tooltip-appear;-webkit-animation-duration:0.1s;animation-duration:0.1s;-webkit-animation-fill-mode:forwards;animation-fill-mode:forwards;-webkit-animation-timing-function:ease-in;animation-timing-function:ease-in;-webkit-animation-delay:0s;animation-delay:0s;}/*!sc*/ .hWlpPn.tooltipped-no-delay:hover::after,.hWlpPn.tooltipped-no-delay:active::after,.hWlpPn.tooltipped-no-delay:focus::after,.hWlpPn.tooltipped-no-delay:focus-within::after{-webkit-animation-delay:0s;animation-delay:0s;}/*!sc*/ .hWlpPn.tooltipped-multiline:hover::after,.hWlpPn.tooltipped-multiline:active::after,.hWlpPn.tooltipped-multiline:focus::after,.hWlpPn.tooltipped-multiline:focus-within::after{display:table-cell;}/*!sc*/ .hWlpPn.tooltipped-s::after,.hWlpPn.tooltipped-se::after,.hWlpPn.tooltipped-sw::after{top:100%;right:50%;margin-top:6px;}/*!sc*/ .hWlpPn.tooltipped-se::after{right:auto;left:50%;margin-left:-16px;}/*!sc*/ .hWlpPn.tooltipped-sw::after{margin-right:-16px;}/*!sc*/ .hWlpPn.tooltipped-n::after,.hWlpPn.tooltipped-ne::after,.hWlpPn.tooltipped-nw::after{right:50%;bottom:100%;margin-bottom:6px;}/*!sc*/ .hWlpPn.tooltipped-ne::after{right:auto;left:50%;margin-left:-16px;}/*!sc*/ .hWlpPn.tooltipped-nw::after{margin-right:-16px;}/*!sc*/ .hWlpPn.tooltipped-s::after,.hWlpPn.tooltipped-n::after{-webkit-transform:translateX(50%);-ms-transform:translateX(50%);transform:translateX(50%);}/*!sc*/ .hWlpPn.tooltipped-w::after{right:100%;bottom:50%;margin-right:6px;-webkit-transform:translateY(50%);-ms-transform:translateY(50%);transform:translateY(50%);}/*!sc*/ .hWlpPn.tooltipped-e::after{bottom:50%;left:100%;margin-left:6px;-webkit-transform:translateY(50%);-ms-transform:translateY(50%);transform:translateY(50%);}/*!sc*/ .hWlpPn.tooltipped-multiline::after{width:-webkit-max-content;width:-moz-max-content;width:max-content;max-width:250px;word-wrap:break-word;white-space:pre-line;border-collapse:separate;}/*!sc*/ .hWlpPn.tooltipped-multiline.tooltipped-s::after,.hWlpPn.tooltipped-multiline.tooltipped-n::after{right:auto;left:50%;-webkit-transform:translateX(-50%);-ms-transform:translateX(-50%);transform:translateX(-50%);}/*!sc*/ .hWlpPn.tooltipped-multiline.tooltipped-w::after,.hWlpPn.tooltipped-multiline.tooltipped-e::after{right:100%;}/*!sc*/ .hWlpPn.tooltipped-align-right-2::after{right:0;margin-right:0;}/*!sc*/ .hWlpPn.tooltipped-align-left-2::after{left:0;margin-left:0;}/*!sc*/ data-styled.g4[id="Tooltip__TooltipBase-sc-17tf59c-0"]{content:"hWlpPn,"}/*!sc*/ masterBranchesTagsGo to fileCodeOpen more actions menu

Folders and filesNameNameLast commit messageLast commit date

Latest commit

History22 Commitsdatadataoutputsoutputs1survey.ipynb1survey.ipynb2satellite.ipynb2satellite.ipynb3ml.ipynb3ml.ipynb4targeting.ipynb4targeting.ipynb5fairness.ipynb5fairness.ipynbREADME.mdREADME.mdhelpers.pyhelpers.pyView all files

Repository files navigation

• README

Togo Targeting Replication Code

Introduction

This repository contains replication code for the paper "Machine Learning and Mobile Phone Data Can Improve Targeting of Humanitarian Assistance" by Emily Aiken, Suzanne Bellue, Dean Karlan, Chris Udry, and Joshua Blumenstock. This readme provides information about the code structure, including where replication code for each figure and table in the paper is located, and information about the data available.

Code Structure

All code is written in Jupyter Notebooks and raw Python, using Python version 3.6. Replication code for all figures and tables that are generated with code is provided. It is assumed that the reader already has access to featurized mobile phone data matched to survey observations (as shown in the synthetic data provided, see section on "data structure" for more information). The scripts are organized into five notebooks as follows. All notebooks call helper functions from the file helpers.py.
#### 1survey.ipynb

Code for all figures and tables that are generated from mainly from survey data, including the calculation of the PMT and asset index, statistics on phone ownership, summary statistics from surveys, and information on weighting and response weights. Replication code for the following tables and figures is included in this notebook:

• Supplementary Figure 1


• Supplementary Figure 2


• Supplementary Figure 3


• Supplementary Figure 4


• Supplementary Figure 10


• Supplementary Figure 11


• Supplementary Table 2


• Supplementary Table 3


• Supplementary Table 4


• Supplementary Table 5


• Supplementary Table 6


• Supplementary Table 10

Code for aggregation of satellite-based wealth estimates and satellite-based population density estimates for high-resolution poverty mapping. Replication code for the following tables and figures is included in this notebook:

• Extended Data Figure 1 Panel A


• Supplementary Figure 5

Code for machine learning from featurized mobile phone data, including matching survey observations to mobile phone records, cross-validation for parameter selection and out-of-sample evaluation, and evaluating feature importances. Replication code for the following tables and figures is included in this notebook:

• Supplementary Figure 7


• Supplementary Figure 8


• Extended Data Table 3


• Extended Data Table 6


• Supplementary Table 13

Code for targeting simulations, including producing targeting tables of all kinds, ROC curves and precision vs. recall curves, and analysis of social welfare. Replication code for the following tables and figures is included in this notebook:

• Figure 1


• Figure 2


• Extended Data Figure 1 Panel B


• Extended Data Figure 4


• Supplementary Figure 9


• Table 1


• Extended Data Table 1


• Extended Data Table 2


• Extended Data Table 4


• Supplementary Table 1


• Supplementary Table 7


• Supplementary Table 8


• Supplementary Table 9

Code for fairness audits of targeting algorithms across potentially sensitive characteristics. Replication code for the following tables and figures is included in this notebook:

• Figure 3


• Extended Data Figure 2


• Extended Data Figure 3

Data Structure

Data files that are publically available are included in this repo; for datasets that are not publically available we have included synthetic (randomly generated) data in the same format and with the same schema. Synthetic data are produced with the notebook `data/generate_synthetic_data.ipynb'. The data files are as follows:

• data/survey2018.csv and data/survey2020.csv: Synthetic data with the schema of the 2018 and 2020 survey datasets, respectively. The 2018 survey is a household survey dataset; the 2020 survey is at the individual level.


• survey_indiv2018.csv: Synthetic individual-level survey data associated with the households from the 2018 survey dataset; used only to calculate statistics on individual-level phone ownership.


• data/features2018.csv and data/features2020.csv: Synthetic data representing a set of featurized mobile phone data. In this file features are randomly generated; in reality they are calculated from raw mobile phone records using open source library bandicoot. Features are provided for a subset of the phone numbers (observations) in each of the 2018 and 2020 synthetic survey datasets.


• data/single_feature2018.csv: Synthetic data representing the "single mobile phone feature" used as a counterfactual targeting method in the paper. The single mobile phone feature is provided for the same set of phone numbers that are associated with full mobile phone featurization.


• data/inferred_home_location2018.csv: Synthetic data representing inferred home locations from mobile phone data. Here home locations are chosen at random; in reality they are inferred from raw mobile phone metadata. Home locations are provided for the same subset of survey observations that are associated with full mobile phone features; they are provided at the prefecture and canton level.


• data/shapefiles: Shapefiles used for poverty mapping. Shapefiles are publicly available from the Humantiarian Data Exchange (https://data.humdata.org/dataset/togo-cod-ab).

• data/satellite: Data for satellite-based poverty mapping.

Running the Code

After installing the necessary packages, the code will run out of the box (using the data files located in the data subfolder). Outputs will be written to the outputs subfolder, divided into output folders for ml, survey, and targeting. Many of the figures and tables are not written to files but rather displayed only in the Jupyter notebooks.{"resolvedServerColorMode":"day"}

About No description, website, or topics provided.

Resources Readme

Uh oh!

There was an error while loading. Please reload this page.Activity

Stars18 stars

Watchers1 watching

Forks10 forks Report repository

ReleasesNo releases published

Packages 0 No packages published

Languages

• Jupyter Notebook91.8%


• Python8.2%

Footer © 2025 GitHub, Inc.

Footer navigation

• Terms


• Privacy


• Security


• Status


• Docs


• Contact


• Manage cookies


• Do not share my personal information You can’t perform that action at this time.