خدمات

مشاوره و تحلیل های آماری  با نرم افزارهای SPSS، SAS، Minitab، Eviews ( در حال حاضر مقدور نیست)

• تحلیل مدلهای خطی عام، رگرسیون لوژستیک و غیره
• تحلیل سری های زمانی، ARIMA، GARCH و SSA

داده کاوی و راه اندازی سیستم های هوش تجاری

مشاوره در خصوص آمارهای رسمی، طراحی برنامه های آماری و غیره   

مشاوره در خصوص مدیریت کیفیت و تضمین کیفیت

خدمات در حوزه آینده پژوهی

• آشنایی با کارکردها و اهداف آینده پژوهی
• روش های آینده پژوهی کمی
• آینده نگاری راهبردی
• پویش محیطی و نحوه استقرار آن
• آشنایی با علایم ضعیف تغییر و کاربرد آن در نوآوری
• تفکر سیستمی و مدل سازی (سیستم های پویا)
• آینده پژوهی فردی و خانوادگی

---------------------------------------------------------------------------------------------------------------------

Data Science Training: 

 

When: 

Wed-Fri, December 2, 3, 4, 2015 - attend 1, 2, or all 3 days 

 

Where: 

Online &

Day 1:

CART®: Modeling with Decision Trees 

Discover the power of predictive modeling, analytics and data mining during this popular introductory seminar, geared toward creating better models using the latest advances in classification and regression technology.

• Explore CART’s unique modeling capabilities while running multiple datasets on both GUI and Non-GUI interfaces, and  discuss the advantages and disadvantages to both. We will explain parts of the display variable importance, summary reporting surrogates and competitors, how the utility handles missing values, and CART's various splitting rules. 

• Discuss CART model evaluation and hybrid models, which includes CART and LOGIT as well as CART and linear regression. We will discuss using relative costs, ROC and cross-validation for model evaluation.
• Introduce the the Powerful Use of Batteries with CART, such as uses for hotspot detection, Battery PRIORS, automatic missing value imputation, and automatic variable binning.

 

Day 2: 

Introduction to Modern Regression Analysis Techniques

Using real-world datasets we will demonstrate Stanford Professor Jerome Friedman's advances in nonlinear, regularized-linear and logistic regression. This workshop will introduce the main concepts behind:

• Generalized PathSeeker (GPS) which includes classic techniques such as ridge and lasso regression, and also adds the new sub-lasso model, as well as intermediate modeling strategies.
• Multivariate Adaptive Regression Splines (MARS), a nonlinear automated regression tool, which automates the model specification search, including variable selection, variable transformation, interaction detection, missing value handling, and model validation. 

 

Day 3:

Advanced Modern Analysis Techniques

This workshop discusses key algorithmic details of Breiman's Random Forests and Friedman's TreeNet, and important extensions to bagging/boosting technology. We will show how the software is used to solve real-world problems, cover theory, discuss what is novel, illustrate how to select an ideal balance between model complexity and predictive accuracy, and show where the software fits in terms of other data mining software: 

• Random Forests is a tree-based procedure making use of bootstrapping and random feature selection.
• TreeNet which builds classification and regression models gradually through a large collection of small trees, each of which improves on its predecessors through an error-correcting strategy.
• Recent developments such as:
  • ISLE, a model compression technology to simplify and speed up complex tree-based ensembles.
  • RuleLearner which reinterprets TreeNet and/or Random Forests tree ensembles, extracting individual segments described by interesting rules. The rules can be combined to yield compressed models, often more accurate than the original ensembles. RuleLearner supports individual-specific and group-specific variable importance rankings and offers dependency plots for model interpretation.
  • ----------------

  • Day 1:

    CART®: Modeling with Decision Trees

    9:00am-10:00am -- Introduction to Predictive Modeling with Decision Trees Using CART

    10:00am-10:15am -- Break

    10:15am-11:15am -- Exploring CART’s Capabilities

    11:15am-11:30am -- Break

    11:30am-12:30pm -- CART Splitting Rules and Tree Building Mechanisms

    12:30pm-1:30pm -- Lunch

    1:30pm-2:30pm -- Model Evaluation and Hybrid Models

    2:30pm-2:45pm -- Break

    2:45pm-3:45pm -- Introduction to the Powerful Use of Batteries with CART

    3:45pm-4:00pm -- Break

    4:00pm-5:00pm -- Loose Ends and Application

     

    Day 2: 

    Introduction to Modern Regression Analysis Techniques

    9:00am-10:00am -- Introduction to Generalized PathSeeker (GPS)

    10:00am-10:15am -- Break

    10:15am-11:15am -- GPS Implementation in SPM

    11:15am-11:30am -- Break

    11:30am-12:30pm -- Regression Trees – A Frequently Overlooked CART Component

    12:30pm-1:30pm -- Lunch

    1:30pm-2:30pm -- Introduction to Multivariate Adaptive Regression Splines (MARS)

    2:30pm-2:45pm -- Break

    2:45pm-3:45pm -- MARS Algorithm

    3:45pm-4:00pm -- Break

    4:00pm-5:00pm -- MARS in Action and Loose Ends and Application

     

    Day 3:

    Advanced Modern Analysis Techniques

    9:00am-10:00am -- Introduction to Ensemble–Based Modeling Techniques

    10:00am-10:15am -- Break

    10:15am-11:15am -- Introduction to Boosting Using Decision Trees

    11:15am-11:30am -- Break

    11:30am-12:30pm -- In-Depth Discussion of TreeNet Theory and Applications

    12:30pm-1:30pm -- Lunch

    1:30pm-2:30pm -- TreeNet in Action

    2:30pm-2:45pm -- Break

    2:45pm-3:45pm -- Interpreting TreeNet Models and Interaction Detection

    3:45pm-4:00pm -- Break

    4:00pm-5:00pm -- Gradient Boosting Unleashed: Combining Models By Linking Engines Together and Loose Ends and Application