Deep Technical & Historical Stock Market Analysis Using OLAP

i have very simple solution, i have pdf file for you, just simply contact me and receive it,

here is some data inside pdf file.

On-Line Analytic Processing
We shall now take up an important class of applications for integrated informa-
tion systems, especially data warehouses. Companies and organizations create
a warehouse with a copy of large amounts of their available data and assign
analysts to query this warehouse for patterns or trends of importance to the or-
ganization. This activity, called OLAP (standing for On-Line Analytic Process-
ing and pronounced \oh-lap"), generally involves highly complex queries that
use one or more aggregations. These queries are often termed OLAP queries
or decision-support queries. Some examples will be given in Section 20.4.1; a
typical example is to search for products with increasing or decreasing overall
sales.
Decision-support queries used in OLAP applications typically examine very
large amounts of data, even if the query results are small. In contrast, common
database operations, such as bank deposits or airline reservations, each touch
only a tiny portion of the database; the latter type of operation is often referred
to as OLTP (On-Line Transaction Processing, spoken \oh-ell-tee-pee").
Recently, new query-processing techniques have been developed that are
especially good at executing OLAP queries eectively. Furthermore, because of
the distinct nature of a certain class of OLAP queries, special forms of DBMS’s
have been developed and marketed to support OLAP applications. The same
technology is beginning to migrate to standard SQL systems, as well. We shall
discuss the architecture of these systems in Section 20.5.
20.4.1 OLAP Applications
A commonOLAP application uses a warehouse of sales data. Major store chains
will accumulate terabytes of information representing every sale of every item
at every store. Queries that aggregate sales into groups and identify signicant
groups can be of great use to the company in predicting future problems and
opportunities.
Example 20.27 : Suppose the Aardvark Automobile Co. builds a data ware-
house to analyze sales of its cars. The schema for the warehouse might be:
Sales(serialNo, date, dealer, price)
Autos(serialNo, model, color)
Dealers(name, city, state, phone)
A typical decision-support query might examine sales on or after April 1, 2001
to see how the recent average price per vehicle varies by state. Such a query is
shown in Fig. 20.19.
Notice how the query of Fig. 20.19 touches much of the data of the database,
as it classies every recent Sales fact by the state of the dealer that sold it.
In contrast, common OLTP queries, such as \nd the price at which the auto
with serial number 123 was sold," would touch only a single tuple of the data.
2
20.4. ON-LINE ANALYTIC PROCESSING 1047
Warehouses and OLAP
There are several reasons why data warehouses play an important role in
OLAP applications. First, the warehouse may be necessary to organize
and centralize corporate data in a way that supports OLAP queries; the
data may initially be scattered across many dierent databases. But often
more important is the fact that OLAP queries, being complex and touch-
ing much of the data, take toomuch time to be executed in a transaction-
processing system with high throughput requirements. OLAP queries of-
ten can be considered \long transactions" in the sense of Section 19.7.
Long transactions locking the entire database would shut down the
ordinary OLTP operations (e.g., recording new sales as they occur could
not be permitted if there were a concurrent OLAP query computing av-
erage sales). A common solution is to make a copy of the raw data in a
warehouse, run OLAP queries only at the warehouse, and run the OLTP
queries and data modications at the data sources. In a common sce-
nario, the warehouse is only updated overnight, while the analysts work
on a frozen copy during the day. The warehouse data thus gets out of date
by as much as 24 hours, which limits the timeliness of its answers to OLAP
queries, but the delay is tolerable in many decision…..