{ "info": { "author": "Bryan S. Graham", "author_email": "bgraham@econ.berkeley.edu", "bugtrack_url": null, "classifiers": [ "Development Status :: 3 - Alpha", "Intended Audience :: Science/Research", "License :: OSI Approved :: MIT License", "Programming Language :: Python :: 2.7", "Topic :: Scientific/Engineering :: Mathematics" ], "description": "ipt: a Python 2.7 package for causal inference by inverse probability tilting\n-----------------------------------------------------------------------------\nby Bryan S. Graham, UC - Berkeley, e-mail: bgraham@econ.berkeley.edu\n\n\nThis package includes a Python 2.7 implementation of the Average Treatment Effect of the \nTreated (ATT) estimator introduced in Graham, Pinto and Egel (2016). The function att() \nallows for sampling weights as well as \"clustered standard errors\", but these features have not\nyet been extensively tested.\n\nAn implementation of the Average Treatment Effect (ATE) estimator introduced in Graham, \nPinto and Egel (2012) is planned for a future update.\n\nThis package is offered \"as is\", without warranty, implicit or otherwise. While I would\nappreciate bug reports, suggestions for improvements and so on, I am unable to provide any\nmeaningful user-support. Please e-mail me at bgraham@econ.berkeley.edu\n\nPlease cite both the code and the underlying source articles listed below when using this \ncode in your research.\n\nA simple example script to get started is::\n\n\t>>>> # Append location of ipt module root directory to systems path\n\t>>>> # NOTE: Only required if ipt not \"permanently\" installed\n\t>>>> import sys\n\t>>>> sys.path.append('/Users/bgraham/Dropbox/Sites/software/ipt/')\n\n\t>>>> # Load ipt package\n\t>>>> import ipt as ipt\n\t\n\t>>>> # View help file\n\t>>>> help(ipt.att)\n\n\t>>>> # Read nsw data directly from Rajeev Dehejia's webpage into a\n\t>>>> # Pandas dataframe\n\t>>>> import numpy as np\n\t>>>> import pandas as pd\n\n\t>>>> nsw=pd.read_stata(\"http://www.nber.org/~rdehejia/data/nsw_dw.dta\")\n\t\n\t>>>> # Make some adjustments to variable definitions in experimental dataframe\n\t>>>> nsw['constant'] = 1 # Add constant to observational dataframe\n\t>>>> nsw['age'] = nsw['age']/10 # Rescale age to be in decades\n\t>>>> nsw['re74'] = nsw['re74']/1000 # Recale earnings to be in thousands\n\t>>>> nsw['re75'] = nsw['re75']/1000 # Recale earnings to be in thousands\n\n\t>>>> # Treatment indicator\n\t>>>> D = nsw['treat']\n\n\t>>>> # Balancing moments\n\t>>>> t_W = nsw[['constant','black','hispanic','education','age','re74','re75']]\n\n\t>>>> # Propensity score variables\n\t>>>> r_W = nsw[['constant']]\n\n\t>>>> # Outcome\n\t>>>> Y = nsw['re78']\n\n\t>>>> # Compute AST estimate of ATT\n\t>>>> [gamma_as, vcov_gamma_ast, study_test, auxiliary_test, pi_eff_nsw, pi_s_nsw, pi_a_nsw, exitflag] = \\\n\t>>>> ipt.att(D, Y, r_W, t_W, study_tilt=True)\n\n\nCODE CITATION\n---------------\nGraham, Bryan S. (2016). \"ipt: a Python 2.7 package for causal inference by inverse probability tilting,\" (Version 0.2.2) \n\t[Computer program]. Available at https://github.com/bryangraham/ipt (Accessed 04 May 2016) \n\t\nPAPER CITATIONS\n---------------\nGraham, Bryan S., Cristine Pinto and Daniel Egel. (2012). \u201cInverse probability tilting for moment condition models \n\twith missing data,\u201d Review of Economic Studies 79 (3): 1053 - 1079\n\nGraham, Bryan S., Cristine Pinto and Daniel Egel. (2016). \u201cEfficient estimation of data combination models by the \n\tmethod of auxiliary-to-study tilting (AST),\u201d Journal of Business and Economic Statistics 31 (2): 288 - 301", "description_content_type": null, "docs_url": null, "download_url": "UNKNOWN", "downloads": { "last_day": -1, "last_month": -1, "last_week": -1 }, "home_page": "http://github.com/bryangraham/ipt", "keywords": "Inverse Probability Tilting (IPT),Auxiliary-to-Study Tilting (AST)", "license": "MIT", "maintainer": null, "maintainer_email": null, "name": "ipt", "package_url": "https://pypi.org/project/ipt/", "platform": "UNKNOWN", "project_url": "https://pypi.org/project/ipt/", "project_urls": { "Download": "UNKNOWN", "Homepage": "http://github.com/bryangraham/ipt" }, "release_url": "https://pypi.org/project/ipt/0.2.3/", "requires_dist": null, "requires_python": null, "summary": "Probability Tilting Methods (IPT) for Causal Inference", "version": "0.2.3" }, "last_serial": 2346973, "releases": { "0.1": [ { "comment_text": "", "digests": { "md5": "4ecfb14165693389dcb8b2b73cc6e06b", "sha256": "112e2289ba14026c9ddec047dbfb2c8b5562f174453c2f2a0c5575bfa1c396d6" }, "downloads": -1, "filename": "ipt-0.1.tar.gz", "has_sig": false, "md5_digest": "4ecfb14165693389dcb8b2b73cc6e06b", "packagetype": "sdist", "python_version": "source", "requires_python": null, "size": 10217, "upload_time": "2016-05-16T21:46:01", "url": "https://files.pythonhosted.org/packages/d3/a5/a3ad9a250168b6a99fd8a40556a49253a766273067d4d90c83c744bf80bd/ipt-0.1.tar.gz" } ], "0.2": [ { "comment_text": "", "digests": { "md5": "b6d790fe2753604da2e67bbfdf7f7de0", "sha256": "9a131bb2ab97ffb00e714a4f72454f10e81ccb0782b89a8fe65ffd6f9ec4d018" }, "downloads": -1, "filename": "ipt-0.2.tar.gz", "has_sig": false, "md5_digest": "b6d790fe2753604da2e67bbfdf7f7de0", "packagetype": "sdist", "python_version": "source", "requires_python": null, "size": 11475, "upload_time": "2016-05-17T18:22:30", "url": "https://files.pythonhosted.org/packages/8b/9e/845ec81a9b62183231738e48ddccf6b589b0fc45ad01e6c5ff2560545216/ipt-0.2.tar.gz" } ], "0.2.1": [ { "comment_text": "", "digests": { "md5": "d569cdf15b3cb87523ec0a5633827f6c", "sha256": "123bcd79f63767c3c17a0bd47c5a0a05264d7c5c193c05e2f08d5c815d1c3bea" }, "downloads": -1, "filename": "ipt-0.2.1.tar.gz", "has_sig": false, "md5_digest": "d569cdf15b3cb87523ec0a5633827f6c", "packagetype": "sdist", "python_version": "source", "requires_python": null, "size": 11487, "upload_time": "2016-05-17T18:45:23", "url": "https://files.pythonhosted.org/packages/e0/e1/acdc8af9f329ebe2f689837489be0891515ae8cce446dd15d503a7a863b1/ipt-0.2.1.tar.gz" } ], "0.2.2": [ { "comment_text": "", "digests": { "md5": "c3731174b62979cef376082c6004606b", "sha256": "a0ce96ab948d650238a88752d8afc3d4bf9c9812ff0cbd205b8ae17361ab4e00" }, "downloads": -1, "filename": "ipt-0.2.2.tar.gz", "has_sig": false, "md5_digest": "c3731174b62979cef376082c6004606b", "packagetype": "sdist", "python_version": "source", "requires_python": null, "size": 13290, "upload_time": "2016-05-30T00:26:02", "url": "https://files.pythonhosted.org/packages/8e/64/c72d880b724ad88fd3e65cb89a39859eb6b912951101f24288c354005aa8/ipt-0.2.2.tar.gz" } ], "0.2.3": [ { "comment_text": "", "digests": { "md5": "2aff9903ddf456a657950ea59ee51288", "sha256": "8ef373c0c1bea823132075fc995f8ea6cb95801b4af09c88009d5bf374a9ecf0" }, "downloads": -1, "filename": "ipt-0.2.3.tar.gz", "has_sig": false, "md5_digest": "2aff9903ddf456a657950ea59ee51288", "packagetype": "sdist", "python_version": "source", "requires_python": null, "size": 13759, "upload_time": "2016-09-16T20:31:17", "url": "https://files.pythonhosted.org/packages/95/b1/09a5b006d074cdfe4aaf86c3539a68b1a5890372b6123084d89edd77c366/ipt-0.2.3.tar.gz" } ] }, "urls": [ { "comment_text": "", "digests": { "md5": "2aff9903ddf456a657950ea59ee51288", "sha256": "8ef373c0c1bea823132075fc995f8ea6cb95801b4af09c88009d5bf374a9ecf0" }, "downloads": -1, "filename": "ipt-0.2.3.tar.gz", "has_sig": false, "md5_digest": "2aff9903ddf456a657950ea59ee51288", "packagetype": "sdist", "python_version": "source", "requires_python": null, "size": 13759, "upload_time": "2016-09-16T20:31:17", "url": "https://files.pythonhosted.org/packages/95/b1/09a5b006d074cdfe4aaf86c3539a68b1a5890372b6123084d89edd77c366/ipt-0.2.3.tar.gz" } ] }