ILLUSTRATION
Contents:
- Totals
- Nominal expenditure
- Current expenditure
- Nominal and real expenditure
- Relative Magnitudes
- Seasonal adjustment
- Comparing over time by using an index
- Patterns within time series data
Totals
When one term will describe the commodity you're dealing with, this is quite a simple choice - you're given the information that the UK imports 'x' thousands of hectolitres of sunflower oil, for example, or that the UK refines 'x' millions of barrels of crude oil for export each year. No problem.
But what if one term causes confusion? Data on the number of new cars registered in the UK may be useful in terms of ascertaining the effect in terms of pollution or congestion, but what if you wanted to analyse the impact of changes in car registrations on demand for petrol or insurance services?
Go to the TimeWeb sample data and get the figure for UK new car registrations in 2000. The series code for the data you are looking for is BCGTAU.
Here, the raw data can't help you analyse in much depth, because it summarises total car registrations. What would help would be to get data for each component category of the composite series. You might want to search for new car registrations according to the engine size, price bracket or geographical location, for example. Also how does this growth compare with GDP growth and income growth?
[Top]
Nominal expenditure
When different commodities are being described, for example changes in the respective demand for flat-packed furniture and mild steel, money is often used as a simple way of expressing totals. Money in this way acts as a unit of account that is broadly understood.
Nominal totals for expenditure on many different commodities can be arrived at simply by multiplying the quantity of the commodity by its price. A total for a longer period is obtained by summing the sub-totals, say for each month of furniture sales, to get the annual nominal total.
[Top]
Current expenditure
Current spending means summations of values expressed in the prices that exist at present. Total current spending on flat-packed furniture is arrived at by multiplying the quantity of each furniture item (tables, chairs, wardrobes and beds, for example) by the current price of these items, then summing the sub-totals.
[Top]
Nominal and real expenditure
Go to the TimeWeb sample data and retrieve the following series:
- IHXTAU - GDP per head current
- IHXWAU - GDP per head constant
Note that the data in the first series are expressed in current prices. These are therefore the nominal GDP per head values. Nominal data can be useful when you want to compare values across different groups, at one point in time.
On many occasions, though, you'll be needing data expressed in real terms.
The data in the second series have been deflated to remove the effects of changes in general prices.
Let's look at the effects of deflating the GDP per head data.
Note the following:
| 1995 | 2000 | |
| Nominal GDP/hd | 12182 | 15668 |
| Real GDP/hd | 12182 | 13744 |
Now try the following question:
You can see that the real data present a value for GDP change that is deflated, accounting for changes in the general level of prices.
[Top]
Relative Magnitudes
On their own, figures don't usually provide much of an insight for the user. To really understand what they mean we have to compare them with data on some related variables. Only then can the scale and importance of the figures be seen.
In the case of total household expenditure, a useful comparison can be made by expressing certain key component variables as percentages of this total figure. This way, the share of the total can be seen and meaningful comparisons made.
Here, all the figures are in real terms, expressed in constant 1995 prices. By making the comparisons, we can see the relative magnitudes involved.
Go to the TimeWeb sample data and retrieve the following series:
ABJRAU Total UK household expenditure in constant 1995 prices
ABZUAU Household expenditure on non-durable goods in constant 1995 prices
ABJZAU Household expenditure on services in constant 1995 prices
AEIWAU Household expenditure on durable goods in constant 1995 prices
| # million 1995 | Total HE | HE non-dur | HE svcs | HE dur |
| Total | 532631 | 227013 | 246369 | 59249 |
|
Exp as % of total HE | 100 | 42.6 | 46.3 | 11.1 |
Households spend a higher proportion of their total expenditure on services than on durable and non-durable goods. Spending on durable goods accounts for only just over one tenth of total expenditure.
[Top]
Seasonal adjustment
Access the following data from the TimeWeb sample:
BAYLAU UK home consumption of beer in 000s hectolitres
BAYMAU UK home consumption of coolers in 000s hectolitres
BAYNAU UK home consumption of wine in 000s hectolitres
Note that these series contain data that is not seasonally adjusted.
When making comparisons, it's important to compare like with like.
In the case of the data series on home consumption of alcoholic drink, all the figures are unadjusted. This means that the data have not been 'smoothed' to remove seasonal fluctuations.
There is more available on seasonal adjustment to help you understand this process.
[Top]
Comparing over time by using an index
Looking at data on different variables that change over time, it can be really hard to get a clear impression of how fast or slowly the variables are changing.
Using a simple index can be very useful in addressing this problem. It can be invaluable when you want to take time series data and plot them graphically.
There is more available on index numbers to help you understand this process.
[Top]
Patterns within time series data
The amount of time between each measurement of a variable in a series is called the frequency or periodicity. Many of the data series on TimeWeb are drawn from sources that publish their data on annual or monthly frequencies.
Annual data can be useful to see 'the big picture', make broad comparisons and look at long-term trends.
Quarterly data help us to see how a variable changes according to the season of the year.
Monthly data series can help us to see more complicated patterns emerge.
In some sectors data with a higher frequency still can be obtained. Exchange rate or stock price movements are recorded on a 'real-time' (as they happen) basis.
When dealing with time series data, you should be aware of the following features of the series:
- seasonal variations
- cycles or variations in business activity resulting from economic booms and slumps
- shocks - unexpected events which affect the whole economy or just the sector in which a firm operates (there is more on this in the glossary which is in the reference section)
- trends - longer term movements caused by structural changes in markets
[Top]
