{
    "info": {
        "author": "Siyuan Li",
        "author_email": "siyuan.li@fitchratings.com",
        "bugtrack_url": null,
        "classifiers": [
            "License :: OSI Approved :: MIT License",
            "Programming Language :: Python :: 3.7"
        ],
        "description": "Function List:\n    update_progress(job_title, progress):\n        show status bar of long process. Operation purpose\n\n    SummarizeDefault(raw_data,default_flag_column,by_vars,plot=False,continuous_var=False,n_bin=10):\n        Summarize default data by different variables\n\n        raw_data: pandas.DataFrame. The dataset contains the default column and groupby variable\n        default_flag_column: String. column name of the default flag in dataset\n        by_vars: String. Column name of the groupby variable\n        continuous_var: Boolean. Default as False. If select as True, function will bucket the variable to do groupby.\n        n_bin: int. If contunuous_var is set as True, function will bucket the variable into \"n_bin\" bins\n        plot: Boolean. Default as False. If select True, function will generate graph\n\n        return a DataFrame\n\n    CalcROCPrep(s, default_flag_data):\n        Preparation for ROC calculation. Not directly into result\n\n    CalcROCStats(s,default_flag_data):\n        Calculation of ROC stats\n        s: pandas.DataFrame. Score generated from regression model\n        default_flag_data: pandas.Dataframe. Default flag data from real transactions.\n\n        return a dictionary of 3 stats\n\n    CalcPredAccuracy(s,default_flag_data,n_bin=10,plot=False,is_default_rate=True):\n        Compare the generated default probability with real default flag\n        s: pandas.DataFrame. Score generated from regression model\n        default_flag_data: pandas.Dataframe. Default flag data from real transactions.\n        n_bin: int. default flag will be grouped into \"n_bin\" bins\n        plot: Boolean. Default as False. If select True, function will generate graph\n        is_default_rate. Boolean. Default as True. Select False only if this function is used to compare two series that are NOT between (0,1)\n\n        return dataframe of bucket score VS actual default\n\n                    CalcModelROC(raw_data,response_var,explanatory_var_1,explanatory_var_2=\"\",explanatory_var_3=\"\",explanatory_var_4=\"\",explanatory_var_5=\"\",explanatory_var_6=\"\",explanatory_var_7=\"\",explanatory_var_8=\"\",explanatory_var_9=\"\",explanatory_var_10=\"\"):\n        Directly generate ROC stats from a given model. Not suggested to use because of explanatory variable number limitation.\n        raw_data: pandas.DataFrame. The dataset contains all variables used in model generation.\n        response_var: string. Column name of response var.\n        explanatory_var_1:string. Column name of response var.\n        explanatory_var_2-10:string.Optional. Column name of response var.\n\n        return dataframe\n\n    CalcWOE_IV(data,default_flag_column,categorical_column,event_list=[1,\"Yes\"],continuous_variable=False,n_bucket=10):\n        Calculate IV of each variable\n        Calculation refer to: https://www.listendata.com/2015/03/weight-of-evidence-woe-and-information.html\n        data: pandas.DataFrame. dataset that contains all variables\n        default_flag_column: String. column name of the default flag in dataset\n        categorical_column:String. column name of the categorical column\n        continuous_var: Boolean. Default as False. If select as True, function will bucket the variable to do groupby.\n        n_bin: int. If contunuous_var is set as True, function will bucket the variable into \"n_bin\" bins\n        event_list: list. The list contains all possibility than can be considered as \"event\"\n\n        return [IV value of certain variable, table of calculation]\n\n    PerformRandomCV(data,response_var,explanatory_var_list=[],k=20,plot=True,sampling_as_default_percentage=False):\n        Do Random Cross Validation of model to check model stability. \n        Return standard deviation of each coefficients.\n        raw_data: pandas.DataFrame. The dataset contains all variables used in model generation.\n        response_var: string. Column name of response var.\n        explanatory_var_list: list. List of all names of explanatory variables\n        k: int. 1/k samples will be eliminated as train set.\n        plot: Boolean. Default as True. Plot std. dev of each coefficient.\n        sampling_as_default_percentage: Boolean. Default as False. If selected as True, sampling will be done as the default rate.\n\n        return std.dev of coefficients during k-time regression\n\n    PerformK_Fold(data,response_var, explanatory_var_list=[],k=10):\n        Perform k-fold cross validation\n        data: pandas.DataFrame. dataset that contains all variables.\n        response_var: string. column name of response variable\n        explanatory_var_list: list of strings. list that contains all explanatory variables\n\n        return array of each fold score\n\n    BestSubsetSelection(data,response_var,explanatory_var_list=[],plot=True):\n        Run best subset selection\n        data: pandas.DataFrame. raw dataset\n        response_var: string. Column name of response var\n        explanatory_var_list: list of string. Names of columns of all explanatory candidates.\n        plot: If selected as True, it will plot the selection steps graph.\n\n        return all combinations' RSS and R_squared\n\n    ForwardStepwiseSelection(data,response_var,explanatory_var_list=[],plot=True):\n        Do forward stepwise selection\n        data: pandas.DataFrame. raw dataset\n        response_var: string. Column name of response var\n        explanatory_var_list: list of string. Names of columns of all explanatory candidates.\n        plot: if selected as True, it will plot the selection steps graph\n\n        return dataframe with AIC,BIC,R_squared of each variable combination\n\n    CalcSpearmanCorrelation(data,var_1,var_2):\n        Calculate spearman Correlation\n        data: pandas.DataFrame. Contains var_1 and var_2\n        var_1,var_2: string. column name of var_1 and var_2\n\n        return float\n\n    CalcKendallTau(data,var_1,var_2):\n        Calculate KendallTau\n        data: pandas.DataFrame. Contains var_1 and var_2\n        var_1,var_2: string. column name of var_1 and var_2\n\n        return float\n\n    CalcSamplingGoodmanKruskalGamma(data,var_1, var_2,n_sample=1000):\n        Calculate GoodmanKruskalGamma estimator with sampling data. Not suggest to use into report as this is just an estimator.\n        data: pandas.DataFrame. Contains var_1 and var_2\n        var_1,var_2: string. column name of var_1 and var_2\n\n        return float\n\n    CalcSomersD(data,var_1,var_2):\n        Calculate Somers' D\n        data: pandas.DataFrame. Contains var_1 and var_2\n        var_1,var_2: string. column name of var_1 and var_2\n\n        return float\n\n    GenerateRankCorrelation(data,var_1,var_2):\n        Calculate all 4 ranked correlation.\n        data: pandas.DataFrame. Contains var_1 and var_2\n        var_1,var_2: string. column name of var_1 and var_2. The order of var_1 and var_2 can not change.\n\n        return float\n\n\n\tOLSForwardStepwise(data,response_var,explanatory_var_list=[],plot=True):\n\t\tPerform forward stepwise selection for OLS regression\n\n",
        "description_content_type": "text/markdown",
        "docs_url": null,
        "download_url": "",
        "downloads": {
            "last_day": -1,
            "last_month": -1,
            "last_week": -1
        },
        "home_page": "https://github.com/siyuan-columbia/FitchValidationTool",
        "keywords": "",
        "license": "MIT",
        "maintainer": "",
        "maintainer_email": "",
        "name": "ValidationTool",
        "package_url": "https://pypi.org/project/ValidationTool/",
        "platform": "",
        "project_url": "https://pypi.org/project/ValidationTool/",
        "project_urls": {
            "Homepage": "https://github.com/siyuan-columbia/FitchValidationTool"
        },
        "release_url": "https://pypi.org/project/ValidationTool/1.0.0/",
        "requires_dist": null,
        "requires_python": "",
        "summary": "Fitch Ratings MVG Validation Tool for criteria models",
        "version": "1.0.0"
    },
    "last_serial": 5784122,
    "releases": {
        "1.0.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "ccf981355ab65400837812929044d128",
                    "sha256": "4ca38f82b88efc5ee15066f0eef1527cc9e228f9ffc4279a9bb12aa5eace38bc"
                },
                "downloads": -1,
                "filename": "ValidationTool-1.0.0-py3-none-any.whl",
                "has_sig": false,
                "md5_digest": "ccf981355ab65400837812929044d128",
                "packagetype": "bdist_wheel",
                "python_version": "py3",
                "requires_python": null,
                "size": 4026,
                "upload_time": "2019-09-05T01:24:19",
                "url": "https://files.pythonhosted.org/packages/67/96/872329d5aaa098fe24bf209b06578d17cf86e9148182f56927bed844c46f/ValidationTool-1.0.0-py3-none-any.whl"
            },
            {
                "comment_text": "",
                "digests": {
                    "md5": "cd08fab2b15e6e75112bb1ad6a59e9bc",
                    "sha256": "8f676885d34a37d72422705e61bc32a5708856e86aaca9a7d852894e01731de7"
                },
                "downloads": -1,
                "filename": "ValidationTool-1.0.0.tar.gz",
                "has_sig": false,
                "md5_digest": "cd08fab2b15e6e75112bb1ad6a59e9bc",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 3302,
                "upload_time": "2019-09-05T01:24:22",
                "url": "https://files.pythonhosted.org/packages/f6/0b/f0489eb3dbeaa0e2f8f79bbb011d922e314aba37dc99e49bc11232493ed4/ValidationTool-1.0.0.tar.gz"
            }
        ]
    },
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "md5": "ccf981355ab65400837812929044d128",
                "sha256": "4ca38f82b88efc5ee15066f0eef1527cc9e228f9ffc4279a9bb12aa5eace38bc"
            },
            "downloads": -1,
            "filename": "ValidationTool-1.0.0-py3-none-any.whl",
            "has_sig": false,
            "md5_digest": "ccf981355ab65400837812929044d128",
            "packagetype": "bdist_wheel",
            "python_version": "py3",
            "requires_python": null,
            "size": 4026,
            "upload_time": "2019-09-05T01:24:19",
            "url": "https://files.pythonhosted.org/packages/67/96/872329d5aaa098fe24bf209b06578d17cf86e9148182f56927bed844c46f/ValidationTool-1.0.0-py3-none-any.whl"
        },
        {
            "comment_text": "",
            "digests": {
                "md5": "cd08fab2b15e6e75112bb1ad6a59e9bc",
                "sha256": "8f676885d34a37d72422705e61bc32a5708856e86aaca9a7d852894e01731de7"
            },
            "downloads": -1,
            "filename": "ValidationTool-1.0.0.tar.gz",
            "has_sig": false,
            "md5_digest": "cd08fab2b15e6e75112bb1ad6a59e9bc",
            "packagetype": "sdist",
            "python_version": "source",
            "requires_python": null,
            "size": 3302,
            "upload_time": "2019-09-05T01:24:22",
            "url": "https://files.pythonhosted.org/packages/f6/0b/f0489eb3dbeaa0e2f8f79bbb011d922e314aba37dc99e49bc11232493ed4/ValidationTool-1.0.0.tar.gz"
        }
    ]
}