Monday, November 16, 2015

Timeline Analysis (Rating: 3.5 out of 5 Stars)

Note: This post represents the synthesis of the thoughts, procedures and experiences of others as represented in the  articles read in advance (see previous posts) and the discussion among the students and instructor during the Advanced Analytic Techniques class at Mercyhurst University in November 2015 regarding Timeline Analysis as a modifier specifically. This modifier was evaluated based on its overall validity, simplicity, flexibility and its ability to effectively use unstructured data.

Timeline analysis is a modifier that can be used to generate a historical context before a decision-making process. The modifier displays events chronologically in a visual manner. Successful applications of this tool can reveal important patterns, key events, or areas on which to focus.

  • Timeline analysis is very easy to conduct
  • Can utilize unstructured data
  • Easy to visually represent for all audiences
  • Identifies key events, thus providing a starting point or key variables for other analytic methods
  • Organizes information for further analysis

  • Timeline analysis is heavily dependent on the ‘binning problem,’ or how many events are grouped into each segment
  • It can be subjective depending on the topic and does not represent causation
  • One must be mindful of biases (i.e. anchoring, confirmation, recency, etc.)
  • Both the visual and the statistical representation can be deceptive
  • Reliant on the chosen starting point, or causal event
  • Once a pattern extracted from a timeline analysis, it may cause an analyst to generate causation from correlations even though that may not be the case

This generic process for conducting timeline analysis was taken from the book, Business and Competitive Analysis: Effective Application of New and Classic Methods by Craig S. Fleisher and Babette E. Bensoussan.
  1. Decide what the timeline will show: major events, market expansions, product introductions, events related to a geographic area, randomly chosen events, and so on.
  2. Make a comprehensive list of events that you want to put on the timeline.  This will require you or others to gather the population of events that are associated with the subject matter being focused on.
  3. Consider how you will choose events to include and exclude from the final timeline.  Not all events will be of equal importance in developing an understanding of the evolution of a firm’s decisions and/or actions.  As such, defining the rules to use for excluding and including events for consideration is important.  The ultimate criteria employed for this task is that these should be based on the client’s critical intelligence needs or topics.
  4. Research and note the specific dates when the events that you want to include occurred. Making a detailed note of your source(s) is a good idea so that you can later verify dates or the details of what transpired.  As well, background documentation should be maintained in separate files for each event, should further examination or inquiries of the events be required.
  5. List the chosen events in a chronology. A chronology is a sequence that starts with the earliest item and ends with the most recent one.  Make special note of the earliest and latest dates that you want to include.  This will also allow you to determine the period of time that their timeline will cover.
  6. Decide what units of time you will use (days, months, quarters, years, decades, and so on) to divide the timeline into segments.  These decisions may be a matter of trial and error.  Calculate the number of segments that your timeline will have.
  7. Draw a line and divide it into the number of equal segments that you figure you will need.
  8. Put the dates on the appropriate segments, from left to right.
  9. Using the chronology that you made of events and dates, figure out where they would fall on your timeline.  Devising a scheme for how you mark and label them is useful.  For instance, you could write certain symbols (for example, $ for acquisitions and * for alliance formations) on the timeline, attach different colored labels, or make a code that refers back to your chronology.
  10. If you do not have room on your timeline to include all of your chronology, cull some of the dates or make a timeline with larger segments (for example, one timeline for events in the firm’s home country and one for events that take place outside its primary market). If your dates can be divided into two or three smaller categories or themes, try making parallel timelines with identical segment sizes.  You can then see how the theme develops, and you can also compare two or more themes at a time.

Personal Application of Technique:
We used a fictitious scenario regarding US and Russian involvement in the Syrian Conflict. A list of fictitious events were provided and students were then asked to analyze the events and determine what the relationships were and what events were important. Several students then shared their key events and we compared. Students also then discussed what the hypothetical question they believed they were answering when conducting their timeline analysis.

Below is the listed fictitious events that students analyzed:

ISIS executes Russian officer in video
Russian troops begin to embed with Syrian Army
US commits division sized force to conflict in Syria
US and Kurdish forces drive ISIS from critical crossroads town
US and Russian ambassadors to the UN engage in heated argument over the actions of their respective militaries in Syria
Hezbollah is effectively driven out of Syria
US Special Forces, working with a moderate Islamic group, are killed by Russian air strike
US lead coalition declares no fly zone over Syrian airspace
US troops killed by members of the Syrian Army, equipped with the latest Russian equipment
ISIS conducts suicide bombing in Moscow, American tourists are among the dead
US F-35 shoots down Russian jet in Syrian no fly zone
US ground forces engage Russian special operations at a distance in Syria
Russia launches ballistic missiles at a US Navy vessel in the Mediterranean sea

The general question students decided they answered was what was the cause and the events leading up to the outbreak of war between Russia and the United States. In this case, the timeline analysis was conducted post-mortem, although usually timeline analysis is ideally conducted pre-mortem.

Trend Analysis and Interpretation: Key Concepts and Methods for Maternal and Child Health Professionals

By: Division of Science, Education and Analysis Maternal and Child Health Bureau



    The public health sector have a long history of monitoring trends and rates of diseases, risk factors, and behavioral tendencies of populations in order to assess current policies and develop new ones in hopes of creating a healthier population. Public health researches usually present data using large populations over a span of many years from national data sets. Time series analysis is also used along side trend analysis to describe the relationship between a risk factor and the outcome, The example the report uses is a study between pollution and hospitalization rates of children with asthma. For the sake of this report, the writer states that the time series analysis discussed focuses on the impact of the trends on hypothesized associations. Public health decision makers are trying to shift from analyzing trends of larger populations, to a smaller group or specific geographic area that would help develop more effective policies and programs. They are also trying to examine trends in indicators of emerging health problems which by definition are only available for short periods of time. Due to this shift in focus, their trend analyses need both descriptive and statistical approaches. However, one must remember that with a smaller sample size, statistical information is often skewed thus reducing analytic confidence.
     This report also discusses the different kinds of error that can occur when taking a statistical approach to trend analysis including random error and sampling error. Also because there might be a lack of data regarding smaller samples or analyses, characterizing the overall shape of a trend could be wrong. This report argues that due to the changing interests in the public health sector, trend analysis must use new strategies when examining data.

Why Do Trend Analysis?

    Health outcomes of a population is most easily understood if their frequency and distribution are described in terms of person, place, and time. Trend analysis can use these terms. Trends may focus on:

  • The overall pattern of change in an indicator over time
  • Comparing one time period to another time period
  • Comparing one geographic area to another
  • Comparing one population to another
  • Making future projections

Preparing to Analyze Trend Data

    A series of conceptual issues must be addressed before analyzing and interpreting trend data regardless of the purpose of the analysis. These issues include:

  • Sample size
    In public health trend analysis is typically conducted at the ecological level meaning observations are time periods and not individuals. The fewer number of time periods, the smaller the sample size which in turn increases one's potential for error. The larger the time period, the more likely it is to precisely identify patterns.
  • Presence of extreme observations or outliers
    The presence of outliers must be analyzed to determine if this incidence is a random occurrence or if it reflects the departure from a general trend which is important for trend analysis.
  • Availability of numerator and denominator data
   "The accuracy of numerator and denominator information over time is also very important in insuring meaningful interpretation of trend data. If both numerator and denominator data for an indicator are available for each time period being studied, then these can be easily analyzed."
  • Confounding (changes over time in factors related to the indicator of interest)
    These changes are especially important to monitor when monitoring smaller populations due to the bigger effect it would have. For example, demographic changes such as a shift in ethnic composition may be more pronounced at the county or community level than in a state or the nation as a whole. Other changes over time that are of concern when conducting trend analysis include changes in laws or public policy that could affect eligibility or access to certain programs, or even cultural trend shifts such as the reduction of substance abuse.

Analysis of Trend Data

     Tables, graphs, and statistical analysis are tools for examining trend data (graphs are especially effective as they provide a visual means of describing a pattern). The first step to assessing a trend is to plot the actual observed numbers by an appropriate time period. This step is important because it establishes the outliers and helps the observer become familiar with both the absolute and relative levels of the numbers being studied.Visual inspection of the data permits the assessment of the overall direction and shape of the trend.
     To make the visual easier to interpret, data transformation and smoothing is possible. For example, putting the rates on a logarithmic scale can 'flatten' the series of rates as seen in this example of infant mortality in Chicago:

      In statistical approaches attempt to 'smooth' data, they aim to increase the stability of the rates by diminishing their jagged appearance. Smoothing can be accomplished by using various forms of averaging, including use of multiple year rates, moving averages, and regression procedures. Public health professionals often use 'collapsing' by combining the numerators and denominators for two or three years of data rather than using the annual rates. This effect increases the stability by increasing the sample size of each point, but also results in a loss of information, leaving less points making it more difficult to portray a pattern. Here is an example of a 'smooth' and collapsed graph:

      Statistical Approaches--There are several statistical approaches in order to test significant trends or different trends:
  • Chi-square test for linear trends--A chi-square test for trend can be obtained by organizing the observed data into a contingency table with one row for each time period and two columns: the first as the numerator, and the second as the denominator.
  • Regression model
    • Ordinary Least Squares (OLS) Regression
      • can be used to model the observed series of rates
    • Poisson Regression
      •  can also be used to model the observed rates, collapsed data, or moving averages
      • accounts for both fluctuation across time and the variability at each time point
      • assumes that errors in modeled observations are independent
    • Time Series Analysis
      • a collection of specialized regression methods that use integrated moving averages and other smoothing techniques and have different assumptions about the error structure of the data
        • Moving average refers to a complex process that incorporates information from past observations and past errors in those observations into the estimation of predicted values
        • Time Series Analysis assumes errors are correlated and can diagnose the precise nature of the correlation and adjust for it
Presentation of Trend Data:

    This report gives general guidelines when presenting this information to a wider audience:


    Although long, this report is very detailed especially when describing the different statistical approaches to trend analysis. For the sake of public health, statistical approaches are better than descriptive ones because public health topics involve larger sample sizes and populations but this is not always the case regarding intelligence questions. I think that in some intelligence cases, the why is very important in determining how a person/country might act which may be the result of a specific occurrence. For its own purposes, I think this report did an excellent job but for our purposes I believe descriptive analysis is also very important when it comes to forecasting.

Saturday, November 14, 2015

 "Autopsy Feature: Graphical Timeline Analysis"
By: Basis Technology


The following is an article about a tool developed to help investigators and intelligence professionals in Timeline Analysis. The tool is a timeline feature as part of the Autopsy software toolkit, which is a software that assists in forensic analysis of computer systems. The Autopsy 3.0.5 will have the feature, thus allowing collection of timeline data for computer forensics.

The system scrapes data from various computer and web sources and arranges them into an activity timeline. In the example, it analyzed the activity concerning a jpeg file of a picture of Osama Bin Laden. It then had a bar chart showing the timeline of the activity concerning the picture. The article then goes into how the tool can be used in other ways. At the macro-level, an investigator can use the tool to see how a specific computer was used. The tool gives a picture of the timeline of the computer. At the micro-level, an investigator can use the tool to analyze the place where a specific breech took place and look at the timeline of events around the breech. In general, this tool gives a timeline autopsy of the computer world.


The tool itself appears to be quite useful, only weakness is that it's only applicable to cyber situations. But general timeline analysis is pretty basic and can be done at a basic level fairly simply. My critique is of the method in general. I see it was useful in criminal investigations when identifying patterns is a vital goal and discovering clues is the primary "requirement". But, when using it to formulate an intelligence estimate I question how effective it is at actually providing a structured reasonably unbiased estimate. It is in my opinion more of a tool to organize information/evidence and then use that timeline to move forward in your analysis. Simply using timeline analysis to make your estimate seems inadequate in most situations.

Friday, November 13, 2015

Event and Timeline Analysis


This is a book chapter. It is like a user's manual of various techniques that can be utilized in strategic and competitive analysis. And the authors describe event and timeline (E&T) analysis as a group of related techniques that display events sequentially in a visual manner. Successful application of this technique can uncover important trends about a firm’s competitive environment and serve as an early-warning function by highlighting when a competitor or another market player is straying from its normal course of behavior.

The authors also assert that this technique can be very beneficial when an analyst is overwhelmed by big data by creating a chronological order. This may facilitate laying out the patterns.

Strengths and Advantages

·      It is a very user friendly technique that is most useful in answering “when is (X event) going to happen?”
·      It is not hard to master this technique. There are also many software applications that can support analysts. And these programs are inexpensive.
·      It is good to use this technique when there are many information that spread over a long period of time.
·      Often used as a planning aid. It can be complementary to other analysis techniques. Therefore, its outputs can be inputs for further analyses.

Weaknesses and Limitations

·      Since there is many information, the technique requires analysts to determine which event should or should not be included in the timeline. This may yield to not putting critical events into the timeline. And this directly affects the findings.
·      Multiple starting points can be generated. Because an analyst may not want to start the timeline to early not to put massive data inputs or he/she may not want to start it too late not to miss any milestone.
·      E&T analysis needs to be done well in advance of key decisions or events. However, the determination of using E&T tends to be after a key event has occurred.
·      It is generally very subjective to determine the causations or distinguishing evidences from assumptions. Any error, especially in causations, may result in misleading patterns.

Application Process of the Technique

1-    Plot the Target Firm’s History of Key Events on a Line (the authors elaborate this step in 10 subsets)
2-    Develop a chronological Table of Events

3-    Develop an Events Matrix

4-    Analyze the event and causal factors


The authors demonstrated finely the strengths, weaknesses and application of the technique. I think this can be a good tool for visualizing or putting the milestone data in an order. However, classifying data as milestone or unimportant is very subjective and therefore, the process is open to errors. And since we all know that correlation doesn’t mean causation, the patterns we generated may inherent false inferences to some extent. And the dimensions of that ‘extent’ will define our results’ forecasting accuracy. To recap, I would be more careful while using timeline analysis with the intention of extracting patterns from that. Other than that, it could be useful in visualizing the existing data.