MatchInterval: Evaluate
Correspondence of
Extreme Climate Events and Cultural Transformations
This program provides as probabilistic evaluation of the
correspondence in time between sets of cultural transformations
and climate records as described and applied in Kintigh and Ingram
(2018). MatchInterval is a Windows program written in
Delphi, an extension of Pascal as implemented in the Embarcadero
RAD Studio XE.
Running the program
Important information on TFQA programs may be found at http://tfqa.com.
See especially,
Program Conventions and
Running TFQA under Windows. To run the program, copy
MatchInterval.exe to a directory that includes the input data
files you wish to evaluate (in CSV Format, as described below).
Having navigated to that folder, when you doubleclick on
MatchInterval.exe the program opens a Windows "Run" Window .
The program will prompt you for information that it needs to run.
Default answers are provided in {curly braces} and can be obtained
by just pressing Enter. Reply Y or N to yes or no questions.
Sample Input file
The program looks for input from two .csv (comma separated
value) files that must be saved in the same directory on your
computer as the .exe file. The CSV Files can be prepared in
Notepad or a similar text editing program or in Excel, saved with
the .csv extension. The transformations file has one line per
cultural transformation with a start and and date, separated by a
comma.The climate extremes file has one line per interval of
climate extremes, again with a start and end date separated by a
comma.
ZuniTrans.csv
1250,1300
1350,1400
ZuniDry.csv
448,464
470,484
512,517
...
1519,1525
Sequence of Program Prompts
┌──────────────────────────────────────────────────────────────────────┐
│ MatchIntervals V2.2 │
│ Monte Carlo Assessment of Correspondence Between 2 Sets of Intervals │
│ │
│ (C) 2013-2016 Keith W. Kintigh │
│ All Rights Reserved │
│ │
│ 2014 East Alameda Drive │
│ Tempe, Arizona 85282 │
└──────────────────────────────────────────────────────────────────────┘
Results File {.TXT} ? Zuni
Name of the output file, to be saved in the same directory.
Fixed (Transformations) Interval File (Start, End) {.CSV} ? ZuniTrans
File: ZUNITRANS.CSV
2 obs & 2 vars; 2 numeric & 0 string
Reply with the name of the file with the periods of
transformations. The program provides some summary information.
Moving (Climate) Interval File (Start, End) {.CSV} ? ZuniDry
File: ZUNIDRY.CSV
31 obs & 2 vars; 2 numeric & 0 string
Reply with the name of the file with the periods of Climate
Extremes. The program provides some summary information.
Start of Earliest Fixed Interval 1250
End of Latest Fixed Interval 1400
Start of Earliest Moving Interval 448
End of Latest Moving Interval 1525
Program report on the input data.
Analysis Interval Start Date {448} ?
Analysis Interval End Date {1525} ?
(Adjusted) Analysis Interval: 448 to 1525
Provide the starting and end dates of the interval you would like
evaluated. These values default to the extreme dates of the
climate data entered but a reduced range can be evaluated.
Number of Random Runs {1000000} ?
Random Generator Seed (0 to set from clock) {0} ?
Enter the number of random runs requested and a random number
generator seed (only if you wish to reproduce a previous run).
Match: [O]verlap, Moving [L]ag Interval; Moving Lag/Fixed [U]ncertainty {U} ?
Impact Time Lag from Beginning of Moving Interval {5} ?
Symmetric Uncertainty around Fixed Interval Start {Y} ?
Fixed Start Uncertainty +/- {5} ?
The program now wants to know what rule you wish to use in
deciding when a climate extreme matches a cultural transformation.
(1) Overlap: any overlap in the intervals. (2) Moving Lag Interval
assumes that there will be a fixed number of lag years from the
start of the climate event to the beginning of a transformation.
(3) Lag with Uncertainty assumes that there is both a lag and some
undertainty associated with the cultural transformation dates.
These options, and the program's operation are described more
fully in the referenced article.
The additional prompts request the Lag and the Uncertainty, if
needed.
Observed Random Probability
Matches Average Random>=Obs
1 0.6103 0.5183
Program Results. The observed number of matches is
the actual number of matches (according to the rule you specified)
between extreme climate events and cultural transformations in the
input data. The Random Average is the average number of matches
(according to the rule) over all random runs between extreme
climate events and cultural transformations. Loosely speaking if
the average number found is in the ballpark of the obserrved, we
should be skecptical of a causal relationship. The
Probability given is the percentage of all random runs in which
the observed number, or more, matches are observed. In this
case, a bit more than half the time a random run produced one (the
observed) or more matches. With a large number of random runs this
analysis may take several seconds to a few minutes depending on
the speed of your computer.
Run Again with Different Decision Rules {N} ?
Run Again with Different Data {N} ?
You can rerun the program if you wish.
Program End
OK to Close Program Window {Y} ?
The program provides this prompt to allow you to examine the
on-screen results. One you reply with Y or Enter it will close
that window and you won’t be able to recover it.
Sample Period Output File
The output file reproduces the input data (truncated here) and
provides the results.
Program MatchIntervals v2.2 - Keith Kintigh - 2020-10-05 8:34:14 AM
================================
All Interval Data
Fixed Interval File: ZUNITRANS.CSV
Moving Interval File: ZUNIDRY.CSV
Fixed Intervals
Start End Years
1250 1300 51
1350 1400 51
Moving Intervals
Start End Years
448 464 17
470 484 15
512 517 6
...
1519 1525 7
================================
Start of Earliest Fixed Interval 1250
End of Latest Fixed Interval 1400
Start of Earliest Moving Interval 448
End of Latest Moving Interval 1525
Analysis Start: 448 End: 1525 Years: 1078
Computation Start: 448 Computation End: 1525 Years: 1078
Computation Start is the start of the moving interval or gap including 448
Computation End is the end of the moving interval or gap including 1525
Years in Moving Intervals: 298 or 27.6% of Computation Interval
Don't start with Gap
Moving intervals must be at least 1 years long to be considered
Only Count 1 Match/Fixed Interval
Random Runs: 1000000
Random Number Generator Seed: -348905323
================================
Intervals Evaluated
Note: Intervals before Analysis Start and End Dates are Ignored
Fixed Intervals
Start End Years
1250 1300 51
1350 1400 51
Moving Intervals Next
Start End Years Gap
448 464 17 5
470 484 15 27
512 517 6 17
...
1519 1525 7
================================
Rule 3 - Lag: 5 Uncertainty -5 : 5
Impact Time Lag & Fixed Interval Uncertainty
Moving Interval Impact Time Lag: 5 from Start of Moving Interval
Impact is Up To 5 Years Before Dated Fixed Interval Start Date and
Impact is Before 5 Years After Dated Fixed Interval Start Date
Match: Fixed 1250-1300 Moving 1250-1254
Observed Random Probability
Matches Matches Rand.>=Obs.
1 0.6103 0.5183
--------------------------------
Program End
References Cited
Kintigh, Keith W. and Scott Ingram
2018
Was the drought really responsible? Assessing statistical
relationships between climate extremes and cultural transitions. Journal
of Archaeological Science 89: 25-31.
DOI:10.1016/j.jas.2017.09.006
Page Last Updated: 4 October 2020
|
