Currencies Analysis Based on Stability: Using Apriori-Algorithm

This paper presents the relative stability based on different currencies. Various currencies have been studied, graphs are plotted with reference to dates on a yearly basis and patterns of the same are analyzed. The stability of currency has been determined by using Apriori-algorithm technique. This algorithm developed is deployed to manoeuvre a continued process until the desired results are achieved.


Introduction
The stability of currency plays a very important role in the economic progress of a nation. It is of paramount importance to the investors. It has been rightly said that stability of a currency is as vital for a country's economy as oxygen is for human beings. It is very important for global economic growth which will boost the fulfilment of human needs. It will also serve as a catalyst to economies and yield positive results to support market economy in a situation where the present tsunami is expected to last for a longer time. Therefore, there has to a line of action which may serve as a cushion for the economies of all countries and allow all governments sufficient scope to handle their economic situation. Even small efforts in this regard will count a lot. This research is based on the insight into currency stability (1, 2).
The Apriori-Algorithm approach determines the selection of most frequently occurring currencies values in a year which is represented by a universal (U) set. The selection from the whole set of various currencies is represented by (S). In this set the most frequently occurring currency with specific dates are noted down and separated from the universal set to yield the result in (X) sets. Specific entries in set (S) are again considered to select more frequently occurring entries among them. This process is continued until desired results are achieved (3-6).
In this paper, section 2 provides the detailed explanation of exchange rates. In section 3 the methodology has been discussed in detail. The simulations have been carried out along with results on the algorithm in section 4. Section 5 provides conclusion and the section 6 elucidates future scope.

Exchange Rates
An exchange rate is the rate at which one currency can be exchanged for another. In other words, it is the value of a country's currency compared to that of the other. Suppose one travels from the US to Egypt. The exchange rate for 1.00 US $ is 5.50 EGP. This means one will get five and a half Egyptian pounds for every US $. Theoretically, identical assets should sell at the same price in different countries, because the exchange rate must maintain the inherent value of one currency against the others (7).

Fixed Rates
There are two ways the price of a currency can be determined against another. A fixed or pegged rate is the rate which the government (central bank) sets and maintains as the official exchange rate. A set price is determined against a major world currency (it is usually the U.S. dollar, but some other major currencies such as euro, yen, pound or a basket of currencies are also considered). In order to maintain the local exchange rate, the central bank buys and sells its own currency in the foreign exchange market in return for the currency to which it is pegged (8).

Floating Rates
Unlike the fixed rate, a floating exchange rate is determined by the private market through supply and demand. A floating rate is often termed "self-correcting", as any differences in supply and demand will automatically be corrected in the market. To be simpler, if demand for a currency is low, its value will decrease, thus making imported goods more expensive and thus stimulating demand for local goods and services. This in return will generate more jobs, and hence an auto-correction would occur in the market. A floating exchange rate is subject to constant change (9).
Another important aspect in this regard is that no currency is wholly fixed or wholly floating. In a fixed regime, market pressures can also influence changes in the exchange rate. When a local currency reflects its true value against its pegged currency, a "black market" which is more reflective of actual supply and demand may develop. A central bank is often forced to revalue/devalue the official rate so that the rate is in line with the unofficial one, thereby halting the activity of the black market. Similarly, in a floating regime, the central bank may also intervene when it is necessary to ensure stability and to avoid inflation. However, it is less often that the central bank interferes in floating regime (10, 11).

Methodology
The methodology adopted consists of the following main steps: a) Judging the stability of currency by using Apriori-Algorithm. b) Choosing data mining tools and selecting data mining techniques to determine the decision graph of stability of currency. c) Testing of algorithm on real market data for authenticity.
Further strategies have been adopted on the basis of the three steps mentioned above (12).

Theoretical Findings
Our findings are based on the following data mining techniques: a) Applying Apriori-Algorithm to currency data to check stability.

b)
Testing on Apriori on real data of Bank of Canada.
After the results are achieved the required graphs are plotted (13, 14).

Currency Apriori Pseudo code
The proposed algorithm pseudo codes are as follows:  Vol. 5, No. 5;May 2010 59 Codes are applied on various data and tested at each stage.

Experimental analysis and findings
The main approaches in this regard are as follows. Figure 1 Stability and security of business depends upon the decrease in the percentage of loss. Figure 1 depicts the stability of the Great Britain Pound (GBP) throughout the year (i.e. 365 days of year 2008). The curves in the graph show how stable the currency is? This is evident from the graph. The variation in graph clearly shows profits and losses of the year. But we have laid stress on the needs of small investors to establish a low scale business with low risk factors [14]. The graph shows a fall below 0.5, especially, in the months of June, July, September and October below 0.5 which indicates instability of GBP. The probabilistic instability per year is 4/12 = 0.333 which means 33% months of the year 2008 are unstable. So investors can not rely on GBP totally.   Figure 3 explains that the graph falling below 0.6. However, this slump is observed only in the month of December. It is negligible during the rest of the year. Its ratio is 1/12 = 0.083, which means about 8.3% months of the year 2008 are unstable as shown in Table 2.

Conclusions
It is concluded that Euro is more stable as indicated by Figures 1-3. This fact has also been verified through the application of Apriori-Algorithm as shown in Table 1. In the case study of conversion, the US Dollar and GBP did not prove to be stable. Further, Table 1 depicts the percentage of probabilistic stability which is 91.27% in the case of Euro. This proves that the investors who invest in Euro for currency conversion businesses, be they small or big, they will not only be safe but will receive good returns on their investments. Further the gold investors are also encouraged to invest in Euro because the buyers and user of gold are less as compared with currency.

Future scope
The algorithm presented in this paper can be applied to software in any desired tools i.e. VB.Net & Oracle or ASP.Net & Microsoft SQL Server. This will enhance the algorithm used in establishing a Micro Currency Exchange data ware house.