HR17 - Repeated change models
Describe repeated fixed and proportional changes with an initial state, generate successive values, and compare their behaviour and assumptions. HR17 includes general iterative processes as well as growth and decay.
General iterative rules
Does multiplying by less than always make a quantity eventually approach zero? A repeated model can also add a fixed amount, so the whole rule matters.
State and update rule
A state is the value at a specified stage. An iterative rule calculates the next state from the current one. Here is the initial state and is the state after completed periods:
Some questions start with instead; use the given indexing rather than assuming the first subscript is zero.
A fixed addition of gives . A repeated percentage change gives , where is the multiplier. Combining proportional change with a fixed addition gives , if the proportional change happens first.
For initial , adding each period gives ; multiplying by gives . The first updated value agrees, but the rules are different.
Worked example 1: growth followed by a fixed deposit
Question. An ideal savings model starts at £. At each month end, the balance increases by , then £ is deposited. Write an initial condition and rule, and find the first three month-end balances.
Plan. Multiply the current balance by , then add . One update represents one month.
Answer. The rule above gives £, £, £.
Check. The first interest amount is £, so . The second is £, giving .
A multiplier below can describe removal of a fraction, with a separate fixed input afterwards.
Worked example 2: proportional loss followed by replenishment
Question. A tank initially contains litres. Each day of the current water is removed, then litres is added. Write the rule and find the next three states.
Plan. Removing leaves , so multiply by before adding .
Answer. , , litres.
Check. The first day removes litres and adds , a net fall of . The second removes and adds , a net fall of .
Record an intermediate table so the output of one update becomes the next input. Keep unrounded values unless the question specifies rounding at each stage; real financial or counting processes may impose such rounding.
Watch out
means “multiply, then add”. Adding first gives , which also multiplies the new input by .
Build and use a fresh rule.
Long-term behaviour
Generating terms shows whether values rise, fall or appear to settle. A fixed point is a state that the rule maps back to itself. Check a proposed fixed point by substitution; a few rounded equal-looking terms alone do not prove exact equality.
Worked example 3: approaching a stable level
Question. Investigate , starting at litres. Does the water approach zero?
Plan. Generate more values and test the candidate level where the daily removal matches the -litre input.
| , litres |
At litres, removal is litres. Substitution gives .
Answer. The values approach litres rather than zero. is a fixed point.
Check. The initial gap above is , then : each gap is multiplied by . Starting at instead gives , approaching the same level from below.
The fixed input changes the long-term level. Compare this with pure proportional changes.
Worked example 4: growth and decay without an input
Question. Two numerical models start at :
Generate three updates and describe their longer-term behaviour if the rules continue unchanged.
Plan. Apply each multiplier repeatedly; distinguish absolute changes from percentage changes.
Answer. keeps growing under this unrestricted model. decreases towards zero but remains positive after every finite number of updates.
Check. Every growth ratio is ; every decay ratio is . The decreases are , not a fixed subtraction of .
Continue this lesson.
You’re previewing NovaLearn without an account. Create a free account to read the rest.
Create free account