Hi guys,
my first post here ever. So far I have been able to find the answers of my questions solved in the web but this time I don't even know how to formulate my issue...
I would like to calculate a number of descriptive statistics of a multiple currencies. The Problem is that I have a file with almost 20.000 data Points containing 30+ currencies. Assume I have 4 coloumns: 1) date, 2) currency Abbreviation, 3) pips (absolute number added ot subtracted fom the spot), and 4) tenors (1m, 2m, 3m, 6m, 12m, 24m, 36m, 48m, and spot) . I actually would like to calc the st. deviation of the pips in order to controll values that fall outside a certain (tbd) number of Standard Deviations (say 1,645 for a 90% confidence interval). To filter out and separate data by currency and date is a massive effort that I cannot aford myself. I was thinking about a formula that seeks out the currency identificator (say, USD) and the corresponding Tenor and pips, and then picking up the values which are available for the time period needed for the calculation of the st Deviation, mean, etc. The time period is determined 1.3.2017 until today. Let's assume this is the base case Scenario.
To further coomplicate this case, there are often "Null" values because data are not every day available for each Tenor or each currency. Another challenge is that the tenors are not equally available for each and every currency; some deliver 24m, others not.
Any idea how to approach this issue?
Thanks,
Evgeni
my first post here ever. So far I have been able to find the answers of my questions solved in the web but this time I don't even know how to formulate my issue...
I would like to calculate a number of descriptive statistics of a multiple currencies. The Problem is that I have a file with almost 20.000 data Points containing 30+ currencies. Assume I have 4 coloumns: 1) date, 2) currency Abbreviation, 3) pips (absolute number added ot subtracted fom the spot), and 4) tenors (1m, 2m, 3m, 6m, 12m, 24m, 36m, 48m, and spot) . I actually would like to calc the st. deviation of the pips in order to controll values that fall outside a certain (tbd) number of Standard Deviations (say 1,645 for a 90% confidence interval). To filter out and separate data by currency and date is a massive effort that I cannot aford myself. I was thinking about a formula that seeks out the currency identificator (say, USD) and the corresponding Tenor and pips, and then picking up the values which are available for the time period needed for the calculation of the st Deviation, mean, etc. The time period is determined 1.3.2017 until today. Let's assume this is the base case Scenario.
To further coomplicate this case, there are often "Null" values because data are not every day available for each Tenor or each currency. Another challenge is that the tenors are not equally available for each and every currency; some deliver 24m, others not.
Any idea how to approach this issue?
Thanks,
Evgeni