Analytics
MIMIC · XMIM · MIM Excel Add-inHistoris' use of temporal interpretation for efficient data storage maximizes research tool retrieval speeds. Because the database utilizes the time-series relationship understanding when performing analysis, it is able to pre-associate all the values within a relation's time series for potential comparison and manipulation. Historis also contains complex algorithms to specifically manipulate date stamps within the database.
The end result is less involvement from the user in understanding data structure, less work in deriving queries, and faster data retrieval.
User-Friendly Query Language
Historis also comes with a unique, English-like query language - MQL - that allows users to ask questions concerning the data without a cryptic programming language.The majority of relational databases utilize Structured Query Language (SQL) for data extraction and reporting. Below are two queries - one in MQL and the other in SQL - each producing the five-day moving average of the closing value of item CL_2007Z when New York City has a high temperature of over 70 degrees in the month of March:
SQL Query:
select pricedata.trans_date as date, pricedata.avgprice as CloseAvg CL_2007Z from
(select a.trans_date, (a.close+ b.close+ c.close + d.close + e.close)/5 as avgprice
from drinyse_cls_d a, drinyse_cls_d b, drinyse_cls_d c, drinyse_cls_d d, drinyse_cls_d e
where a.vsid = 1506568
and a.trans_date -1 = b.trans_date and a.trans_date -2 = c.trans_date
and a.trans_date -3 = d.trans_date and a.trans_date -4 = e.trans_date
and a.vsid = b.vsid and a.vsid = c.vsid
and a.vsid = d.vsid and a.vsid = e.vsid
and a.status_flag = b.status_flag
and a.status_flag = c.status_flag
and a.status_flag = d.status_flag
and a.status_flag = e.status_flag
and a.status_flag = 'H' ) pricedata,
(select trans_date, hightemp from ncep_hlp_d where vsid = 175263 and status_flag = 'H'
) wthrdata
where pricedata.trans_date = wthrdata.trans_date and wthrdata.hightemp > 70
and to_char(pricedata.trans_date, 'MM') = '03' order by pricedata.trans_date;
MQL Query:
SHOW
5 day average of Close of CL_2007Z
WHEN
hightemp of New.York.City.NY is more than 70
AND
Date is March
select pricedata.trans_date as date, pricedata.avgprice as CloseAvg CL_2007Z from
(select a.trans_date, (a.close+ b.close+ c.close + d.close + e.close)/5 as avgprice
from drinyse_cls_d a, drinyse_cls_d b, drinyse_cls_d c, drinyse_cls_d d, drinyse_cls_d e
where a.vsid = 1506568
and a.trans_date -1 = b.trans_date and a.trans_date -2 = c.trans_date
and a.trans_date -3 = d.trans_date and a.trans_date -4 = e.trans_date
and a.vsid = b.vsid and a.vsid = c.vsid
and a.vsid = d.vsid and a.vsid = e.vsid
and a.status_flag = b.status_flag
and a.status_flag = c.status_flag
and a.status_flag = d.status_flag
and a.status_flag = e.status_flag
and a.status_flag = 'H' ) pricedata,
(select trans_date, hightemp from ncep_hlp_d where vsid = 175263 and status_flag = 'H'
) wthrdata
where pricedata.trans_date = wthrdata.trans_date and wthrdata.hightemp > 70
and to_char(pricedata.trans_date, 'MM') = '03' order by pricedata.trans_date;
MQL Query:
SHOW
5 day average of Close of CL_2007Z
WHEN
hightemp of New.York.City.NY is more than 70
AND
Date is March
Which query would you rather input? As you can see, the MQL query is simpler, easier to read, contains less text, and requires a very limited understanding of the database.
Additionally, MQL's understanding of time-series relationships enables the Historis database to perform much more complex analysis than SQL, which doesn't pre-associate any given value with another.
Analytic Tools
Historis provides several tools for analysis - XMIM, MIMIC and EMIM Excel Add-in - that can be leveraged to draw conclusions and otherwise make sense of time-series data.Historis also provides links to popular analytical tools S-PLUS and MATLAB and integrates easily with most existing data analytics packages.