Conference PaperPDF Available

Interactive Simulation and Visualization of Long-Term, ETF-based Investment Strategies


Abstract and Figures

Personal, long-term investment products, especially ones for retirement savings, require thorough understanding to use them profitably. Even simple savings plans based on exchange-traded funds(ETFs) are subject to many variables and uncertainties to be considered for expected and planned-upon returns. We present aninteractive simulation of an ETF-based savings plan that combinesforecasts, risk awareness, taxes and costs, inflation, and dynamicinflows and outflows into a single visualization. The visualizationconsists of four parts: a form-fill interface for configuration, a savings and payout simulation, a cash flow chart, and a savings chart. Based on a specific use case, we discuss how private investors canbenefit from using our visualization after a short training period.
Content may be subject to copyright.
Interactive Simulation and Visualization of
Long-Term, ETF-based Investment Strategies
Martin Büßemeyer
Hasso Plattner Institute, Digital Engineering Faculty
University of Potsdam, Germany
Daniel Limberger
Hasso Plattner Institute, Digital Engineering Faculty
University of Potsdam, Germany
Willy Scheibel
Hasso Plattner Institute, Digital Engineering Faculty
University of Potsdam, Germany
Jürgen Döllner
Hasso Plattner Institute, Digital Engineering Faculty
University of Potsdam, Germany
Personal, long-term investment products, especially ones for retire-
ment savings, require thorough understanding to use them prof-
itably. Even simple savings plans based on exchange-traded funds
(ETFs) are subject to many variables and uncertainties to be con-
sidered for expected and planned-upon returns. We present an
interactive simulation of an ETF-based savings plan that combines
forecasts, risk awareness, taxes and costs, ination, and dynamic
inows and outows into a single visualization. The visualization
consists of four parts: a form-ll interface for conguration, a sav-
ings and payout simulation, a cash ow chart, and a savings chart.
Based on a specic use case, we discuss how private investors can
benet from using our visualization after a short training period.
Human-centered computing Information visualization.
interative visualization, nancial visualization, 401(k) plan visualiza-
tion, ETF savings plan visualization, retirement savings calculator
ACM Reference Format:
Martin Büßemeyer, Daniel Limberger, Willy Scheibel, and Jürgen Döll-
ner. 2021. Interactive Simulation and Visualization of Long-Term, ETF-
based Investment Strategies. In The 14th International Symposium on Vi-
sual Information Communication and Interaction (VINCI ’21), September
6–8, 2021, Potsdam, Germany. ACM, New York, NY, USA, 5 pages. https:
Pension entitlement is currently falling continuously due to the
demographic change [
]. To make up for a gap in personal
retirement provision, one can resort to oers from nancial institu-
tions. For example, Riester in Germany [
] or the more widely
known 401(k) plans in the USA [
]. The underlying strategies
are often actively managed and produce unnecessary costs, while
Permission to make digital or hard copies of all or part of this work for personal or
classroom use is granted without fee provided that copies are not made or distributed
for prot or commercial advantage and that copies bear this notice and the full citation
on the rst page. Copyrights for components of this work owned by others than the
author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or
republish, to post on servers or to redistribute to lists, requires prior specic permission
and/or a fee. Request permissions from
VINCI ’21, September 6–8, 2021, Potsdam, Germany
©2021 Copyright held by the owner/author(s). Publication rights licensed to ACM.
ACM ISBN 978-1-4503-8647-0/21/09. . . $15.00
-2K €
-1K €
0K €
1K €
2K €
3K €
4K €
-50K €
0K €
50K €
100K €
150K €
200K €
Figure 1: Our visualization of a plan with a dynamic e100
monthly investment split into three accumulating ETFs.
The cash ow chart (top) depicts the monthly investments
and payouts. The depot chart (bottom) shows the simulated
forecast of the savings, taxes (red), costs, and ination (ma-
genta) over a time-span of 40 (saving) plus 30 (payout) years.
This shows, i.a., a dynamic payout of e800 with the start of
the pension is reasonably sustainable for at least 20 years.
passive strategies, tend to win over active strategies in the long
run [
]. This is one of the reasons why passive investment strate-
gies based on long-term saving plans using exchange-traded funds
(ETFs) have become quite popular. more attraction in other coun-
tries. Since most direct banks and brokers provide opportunities for
regular, small investments in ETFs with reduced costs, it is fairly
easy to minimize costs and administrative expenses in order to max-
imize savings. The ETF investment system itself, however, is rather
complicated and requires stock market literacy [
]. Therein lies a
fundamental conict of interest, as forecasts from brokers or banks
themselves are not designed to be transparent to the customer. In
order to maximize incentives to create a depot everything is kept
as simple as possible. This leads to over-simplied communication
of costs, taxes, and idealized forecasts [
]. The main contributions
of this work are
a savings plan simulation that takes growth, dividends, taxa-
tion, and transaction costs into account, and
a visualization using a cash ow chart and a savings chart
accounting for ination, taxes, costs, etc.
VINCI ’21, September 6–8, 2021, Potsdam, Germany M. Büßemeyer, D. Limberger, W. Scheibel, J. Döllner
“The most straightforward (and generic) approach to enhancing
the value of an unknown outcome is to communicate concrete
consequences” [
]. Thus, self-ecacy with regard to nancial
decision-making improves the likelihood that individuals follow
the retirement strategy [
]. “[A] recent study found that less than
25% of [American] individuals over the age of 50 succeeded at saving
for retirement, often because they were unaware of fundamental
economic concepts such as interest rate or ination” [17].
We propose an interactive savings simulation and target non-
experts to support decision-making on ETF-based saving plans (Fig-
ure 1). We enable interactive examination of various aspects of a
savings plan by making all calculations in real time and provide
immediate visual feedback. This allows users to (1) understand the
interactions of parameters and risks during savings and payout
phase, and to (2) identify scenarios that t their own needs. Our
open source prototype focuses on Germany’s tax law (could be
adapted to other countries) and is available at
“Financial literacy empowers people to craft their nances in a way
that they are able to manage their everyday expenses, maintain
an emergency fund, plan for children’s education and prepare for
their swift post-retirement years” [
]. This becomes more important
since employers and the federal government forward the respon-
sibility to the individual resident [
]. This literacy is understood
in a broader sense and “[v]isualizations may be [...] [a] good ap-
proach for overcoming processing diculties because they shift
information processing to the perceptual system, improving under-
standing of the concept presented and allowing decision-makers
to quickly learn from trends and patterns in the data” [
]. For
example, visualization is used to get an overview on whole markets
and publicly available nance data [
], or underlying mechanics
and consequences [
]. Visualization has also become available for
a more personal use [
] and was used to assess and communicate
risks for savings [
]. for growth. We were surprised to see how few
tools for forecasting and visualizations are currently available for
end users. Those available, however, usually kept relevant factors
like costs, ination, and taxes out of consideration [
]. Similar is-
sues are present in user driven visualizations that generalize across
country borders and thus cost and taxation models [
] though
they are more transparent in their simulations.
There are several statistical models used for predicting stock
markets. One prominent model is ARIMA. Though research has
shown that ARIMA only works in some cases for forecasting short-
term periods [
]. Moreover, we are aiming for long-term
forecasting periods. We can see that the market evolved in an
exponential fashion during the last decades [
]. Thus, simpler
regression models seem sucient for our long-term forecasting.
Our simulation provides access to several ETFs, uses historical data
for forecasting, and takes costs and German taxes into account.
Source Data. We use historical data from Alphavantage to fore-
cast prices. Their API provides the average monthly closing value
for the last ve years of many well-known ETFs. Similar services
are available from Yahoo Finances and xignite.
German Costs and Taxation Model. German citizen have a yearly
exemption amount that can be used on investment gains to save
taxes. The tax amount is calculated by multiplying the gain by the
partial exemption rate of 0
7. This prevents double taxation that
would occur since the ETF owner has already paid taxes. The tax
rate is composed of corporate income tax and the solidarity contri-
bution, which add up to 0
26375%. In Germany, the FIFO principle
is prescribed for the sale of shares: the oldest shares are sold rst.
Since these are likely to have the highest prot, taxes due at the be-
ginning of the payout period will be highest and decrease over time.
Accumulating and distributing ETFs dier in the way dividends are
distributed and taxed. Compared to distributing ETFs, accumulating
ETFs reinvest the distribution automatically. Instead of a dividend,
the investor receives additional shares and no taxes are due. In 2018,
the Vorabpauschale (advance at taxes) was introduced as part of an
investment tax reform. The taxes that need to be paid for a year’s
prot are limited by the expected prot. The expected prot is
calculated by multiplying the amount invested by the base interest
rate and is discounted proportionately for each month of the year
that has already passed. Finally, the money distributed for the year
is deducted from the resulting amount. Thus, there are eectively
no taxes for distributing ETFs. Other costs arise from brokerage
fees, usually a mix of xed amounts and percentage costs.
Growth Model. Statistical models like MA and ARIMA are used
for short-term forecasts. Since we have to forecast far into the
future, these models are not suitable. Instead, models capturing
general price development of recent decades seem to be sucient.
Even though, prices have continued to rise exponentially [
], we
decided to be pessimistic and assume linear model inspired by [
in our simulation to avoid overly optimistic forecasts. This can be
easily replaced by an exponential model, if desired, or adjusted
directly by means of a condence score on the price development.
Savings. The savings phase starts with an optional starting capi-
tal and continues with a static or dynamic monthly amount. New
shares are calculated by dividing the investment amount without
costs by the current price. The Vorabpauschale and payout of money
take place in December. If a distributing strategy is used, taxes on
returns are considered. The payout is then reinvested without ad-
ditional costs and the Vorabpauschale for this year is calculated.
Algorithm 1: Pseudo code of the calculation for a saving month.
1𝑑𝑖𝑠𝑡𝑟 𝑖𝑏𝑢𝑡 𝑖𝑜𝑛𝑠 0
2𝑐𝑜𝑠𝑡 𝑠, 𝑖𝑛 𝑣𝑒𝑠𝑡 𝑚𝑒𝑛𝑡 𝑐 𝑎𝑙𝑐𝑢𝑙𝑎𝑡 𝑒𝐶𝑜𝑠 𝑡𝑠 (𝑖𝑛𝑣𝑒𝑠𝑡𝑚𝑒𝑛𝑡 )
3𝑛𝑒𝑤𝑆ℎ𝑎𝑟𝑒𝑠 𝑖𝑛𝑣𝑒𝑠𝑡 𝑚𝑒𝑛𝑡 /𝑐𝑢𝑟𝑟𝑒𝑛𝑡𝑃 𝑟𝑖𝑐𝑒
4if 𝑐𝑢𝑟𝑟𝑒𝑛𝑡𝑀 𝑜𝑛𝑡 == 𝐷𝑒𝑐 𝑒𝑚𝑏𝑒𝑟 then
5if isDistributionModel then
6𝑑𝑖𝑠𝑡𝑟 𝑖𝑏𝑢𝑡 𝑖𝑜𝑛𝑠 𝑑 𝑖𝑠𝑡 𝑟𝑖𝑏𝑢𝑡𝑖𝑜𝑛
7𝑡𝑎𝑥 𝑒𝑠, 𝑑 𝑖𝑠𝑡 𝑟𝑖𝑏𝑢 𝑡𝑖𝑜 𝑛, 𝑡𝑎𝑥 𝐹 𝑟𝑒 𝑒
𝑎𝑝𝑝𝑙 𝑦𝑇 𝑎𝑥𝑒𝑠 (𝑑 𝑖𝑠𝑡 𝑟𝑖𝑏𝑢𝑡 𝑖𝑜𝑛, 𝑡 𝑎𝑥 𝐹𝑟 𝑒𝑒)
8𝑛𝑒𝑤𝑆ℎ𝑎𝑟𝑒𝑠 𝑛𝑒 𝑤𝑆ℎ𝑎𝑟 𝑒𝑠 +𝑑 𝑖𝑠𝑡 𝑟𝑖𝑏𝑢𝑡𝑖𝑜𝑛/𝑐𝑢𝑟 𝑟 𝑒𝑛𝑡 𝑃𝑟 𝑖𝑐 𝑒
9𝑒𝑥 𝑝𝑒𝑐𝑡𝑒𝑑 𝐺𝑎𝑖𝑛 𝑐 𝑎𝑙𝑐𝑢𝑙 𝑎𝑡𝑒𝐸𝑥 𝑝𝑒 𝑐𝑡 𝑒𝑑𝐺 𝑎𝑖𝑛 ()
10 𝑣𝑜𝑟 𝑎𝑏𝑃 𝑐𝑙 𝑎𝑚𝑝 (𝑔𝑎𝑖𝑛, 0,𝑒 𝑥𝑝 𝑒𝑐𝑡𝑒 𝑑𝐺𝑎𝑖𝑛 𝑑𝑖𝑠 𝑡𝑟 𝑖𝑏𝑢𝑡 𝑖𝑜𝑛𝑠)
11 𝑡𝑎𝑥 𝑒𝑠 𝑡 𝑎𝑥𝑒𝑠 +𝑎𝑝𝑝𝑙 𝑦𝑇 𝑎𝑥𝑒𝑠 (𝑚𝑎𝑥 (0, 𝑣𝑜𝑟 𝑎𝑏𝑃 𝑡𝑎𝑥 𝐹 𝑟 𝑒𝑒 ))
12 𝑖𝑛 𝑓 𝑙𝑎𝑡𝑖𝑜 𝑛 𝑡 𝑜𝑡 𝑎𝑙𝑉 𝑎𝑙𝑢𝑒 𝑡𝑜 𝑡𝑎𝑙𝑉 𝑎𝑙𝑢 𝑒 ∗ (10.02)𝑦𝑒𝑎𝑟 𝑠𝑃 𝑎𝑠𝑠𝑒𝑑
13 return newShares, costs, taxes, ination
Interactive Simulation and Visualization of Long-Term, ETF-based Investment Strategies VINCI ’21, September 6–8, 2021, Potsdam, Germany
Algorithm 2: Pseudo code of the calculation for a payout month.
1𝑟𝑒𝑐𝑒𝑖 𝑣𝑒𝑑 𝑃𝑎 𝑦𝑜𝑢𝑡 0
2if 𝑐𝑢𝑟𝑟𝑒𝑛𝑡𝑀 𝑜𝑛𝑡 == 𝐷𝑒𝑐 𝑒𝑚𝑏𝑒𝑟 then
3if isDistributionModel then
4𝑑𝑖𝑠𝑡𝑟 𝑖𝑏𝑢𝑡 𝑖𝑜𝑛𝑠 𝑑 𝑖𝑠𝑡 𝑟𝑖𝑏𝑢𝑡𝑖𝑜𝑛
5𝑡𝑎𝑥 𝑒𝑠, 𝑑 𝑖𝑠𝑡 𝑟𝑖𝑏𝑢 𝑡𝑖𝑜 𝑛 𝑎𝑝𝑝𝑙 𝑦𝑇 𝑎𝑥𝑒𝑠 (𝑑𝑖 𝑠𝑡 𝑟𝑖𝑏𝑢𝑡𝑖𝑜𝑛)
6𝑝𝑎𝑦𝑜𝑢𝑡 𝑝𝑎 𝑦𝑜𝑢𝑡 𝑑𝑖 𝑠𝑡 𝑟𝑖𝑏𝑢𝑡𝑖𝑜𝑛
8𝑛𝑒𝑤𝑆ℎ𝑎𝑟𝑒𝑠 𝑑𝑖𝑠𝑡𝑟 𝑖𝑏𝑢𝑡𝑖𝑜 𝑛/𝑐𝑢𝑟 𝑟𝑒 𝑛𝑡𝑃 𝑟 𝑖𝑐𝑒
9for oldestInvestment inv: until payout == 0 do
10 𝑔𝑎𝑖𝑛 𝑐𝑎𝑙𝑐𝑢𝑙𝑎𝑡 𝑒𝐺𝑎𝑖𝑛 (𝑖𝑛𝑣)
11 𝑛𝑒𝑤𝑇 𝑎𝑥𝑒 𝑠, 𝑛𝑒 𝑤𝑃𝑎 𝑦𝑜𝑢𝑡 , 𝑡𝑎𝑥 𝐹𝑟 𝑒𝑒 𝑎𝑝𝑝𝑙 𝑦𝑇 𝑎𝑥𝑒𝑠 (𝑔𝑎𝑖𝑛, 𝑡 𝑎𝑥𝐹 𝑟 𝑒𝑒)
12 𝑝𝑎𝑦𝑜𝑢𝑡 𝑝𝑎 𝑦𝑜𝑢𝑡 𝑖𝑛 𝑣
13 𝑟𝑒𝑐𝑒𝑖 𝑣𝑒𝑑 𝑃𝑎 𝑦𝑜𝑢𝑡 𝑟 𝑒𝑐 𝑒𝑖 𝑣𝑒𝑑 𝑃𝑎 𝑦𝑜𝑢𝑡 + (𝑖𝑛𝑣 𝑛𝑒 𝑤𝑇𝑎𝑥𝑒𝑠 )
14 𝑡𝑎𝑥 𝑒𝑠 𝑡 𝑎𝑥𝑒𝑠 +𝑛𝑒𝑤𝑇 𝑎𝑥 𝑒𝑠
15 if isDistributionModel then
16 𝑒𝑥 𝑝𝑒𝑐𝑡𝑒𝑑 𝐺𝑎𝑖𝑛 𝑐 𝑎𝑙𝑐𝑢𝑙 𝑎𝑡𝑒𝐸𝑥 𝑝𝑒 𝑐𝑡 𝑒𝑑𝐺 𝑎𝑖𝑛 ()
17 𝑣𝑜𝑟 𝑎𝑏𝑃 𝑐𝑙 𝑎𝑚𝑝 (𝑔𝑎𝑖𝑛, 0,𝑒 𝑥𝑝 𝑒𝑐𝑡𝑒 𝑑𝐺𝑎𝑖𝑛 𝑑𝑖𝑠 𝑡𝑟 𝑖𝑏𝑢𝑡 𝑖𝑜𝑛𝑠)
18 𝑡𝑎𝑥 𝑒𝑠 𝑡 𝑎𝑥𝑒𝑠 +𝑎𝑝𝑝𝑙 𝑦𝑇 𝑎𝑥𝑒𝑠 (𝑚𝑎𝑥 (0, 𝑣𝑜𝑟 𝑎𝑏𝑃 𝑡𝑎𝑥 𝐹 𝑟 𝑒𝑒 ))
19 𝑐𝑜𝑠𝑡 𝑠, 𝑟 𝑒𝑐𝑒 𝑖𝑣𝑒𝑑 𝑃𝑎 𝑦𝑜𝑢𝑡 𝑐 𝑎𝑙𝑐𝑢𝑙 𝑎𝑡𝑒𝐶𝑜 𝑠𝑡𝑠 (𝑝𝑎 𝑦𝑜𝑢𝑡 )
20 𝑖𝑛 𝑓 𝑙𝑎𝑡𝑖𝑜 𝑛 𝑡 𝑜𝑡 𝑎𝑙𝑉 𝑎𝑙𝑢𝑒 𝑡𝑜 𝑡𝑎𝑙𝑉 𝑎𝑙𝑢 𝑒 ∗ (10.02)𝑦𝑒𝑎𝑟 𝑠𝑃 𝑎𝑠𝑠𝑒𝑑
21 return newShares, receivedPayout, costs, taxes, ination
Finally, the eectively lost amount of money caused by ination is
calculated. This process is shown in Algorithm 1.
Payouts. During the payout phase, a specied monthly payout
is withdrawn from the depot. For the distributing variant, payouts
in December are not reinvested directly, but used for the monthly
payout rst. Next, an iteration over the past investments is done
according to the FIFO principle in order to withdraw the required
amount. For each past investment, the gain is calculated and the
occurring taxes (including the Vorabpauschale in December) are
deducted from the amount. Finally, the occurring broker costs and
the ination are calculated. The process is shown in Algorithm 2.
The visualization features two charts, the cash ow and detail view,
which are synced via a time-axis (x-axis) The switch from saving to
payout phase is indicated by a vertical line. The end of the payout
phase fades out to indicate inaccuracies in life expectancy. The
actual data at a certain time is printed on the right side when
hovering over that point in time. The hovering is synced across
Saving Phase
Payout Phase
ETF2 ... ETFn-1
Savings Total
Figure 2: An explosion view depicting the composition of
our depot view: The upper part consists of 𝑛ETFs with their
values stacked on top of each other. An additional line on
top indicates the total saving. The negative space consists of
costs followed by taxes and ination. The x-axis splits posi-
tive and negative space and serves as semantic separator.
-100K €
0K €
100K €
200K €
300K €
400K €
Figure 3: The detail view when the condence display option
is activated. The standard stacked areas are more transpar-
ent and the condence area is placed over it. The area dis-
plays the possible range of the total depot value. The area
borders are enhanced by a green (maximal condence) and
red (minimal condence) line. The used condence interval
is 0%-100%. The main blue line displays the condence of
the current conguration (30%). Other lines are added arti-
cially, displaying condences in 10% steps upwards.
both charts and is displayed via a vertical dashed line. Next, the
main user interface contains all parameterization options and is
placed on the left. The whole prototype is depicted in Figure 4.
Cash Flow Overview. The cash ow chart visualizes the actual
account movements using a bar chart. The red bars indicate the in-
vestment, whereas the green bars represent the payout. The payout
already includes all fees and taxes. Each bar aggregates a certain
congurable time period, e.g., a month or a quarterly period.
Detail View. The detail view displays the depot value develop-
ment and all associated negative eects: costs, taxes, and ination.
A stacked area chart (Figure 2) with a negative and positive part
is used. The positive part displays the value development of the
ETFs. Thereby, each ETF consists of two stacks. The lighter color
shows the investment and the darker one is for the revenue. An
additional line emphasizes the total value of the depot. The negative
part consists of the accumulated costs (purple) and the accumulated
taxes (red). Additionally, the ination is added as a visual cue. On
the bottom of this chart, all actual values are listed in text form.
They display the dierence in the respective value to the previous
parameter conguration in respect to the hovered point in time.
Moreover, a condence area can be optionally added over the exist-
ing stacked areas. In this case, the borders of the condence area
are the total depot value that would incur when the minimum and
maximum condence would be used. The middle line displays the
currently selected condence and is identical to the total value line
when the condence area is not displayed (Figure 3).
Parameterization. Parameters are divided in the categories money,
time, cost, and visualization. In the money options the starting cap-
ital, monthly investment, and payout can be set. Investment and
payout amounts can be set to dynamically increase on a yearly
basis. The cost section gives control to the costs used for the saving
and payout phase. Furthermore, the time options specify the length
of both phases by means of the current age, life expectation, and
years left until retirement. The visualization options include ETF
selection and condence conguration used for forecasting.
VINCI ’21, September 6–8, 2021, Potsdam, Germany M. Büßemeyer, D. Limberger, W. Scheibel, J. Döllner
Figure 4: A complete overview of the existing prototype. A
parameter pane divided into money, time, cost, and visual-
ization options is on the left. The cash ow chart (top) dis-
plays the yearly investments and payouts. The depot chart
(center) displays the forecast including costs, taxes, and in-
ation. Details w.r.t. to the year pointed at are displayed on
right next to each chart. An agglomeration of all values is
displayed in a tabular fashion (bottom).
In this section, we (1) discuss the results of our prototypical simula-
tion and visualization and (2) evaluate it by applying it to a certain
use case and show the usefulness of the visualization.
Use Case: Family Retirement Savings. Let’s imagine a young fam-
ily with one child. We assume a starting capital of
1000, a monthly
investment of
100 and payout of
1000. In order to compen-
sate the ination both, the investment and payout increases by 1%
per year. Additionally, the tax-free amount for a married couple is
1602 and the accumulating version of the ETFs shall
be used. Moreover, the couple is around 25 years old, will retire
in 43 years and has a life expectation of 80 years. Their invest-
ment is equally spread across two standard ETFs the S & P 500 and
MSCI EM. Furthermore, the family is assumed to already have an
account at the bank ING which has relative high investment costs
4.99 xed costs and 0
25% percentage costs. The condence
is set to 100% to minimize expectation. The simulation predicts a
depot value of
161000 with overall costs of
0 taxes, and
60000 eectively lost to ination at the end of the saving phase.
Their depot lasts until 2084 which is one year earlier before the
expected lifetime. Until that date,
8000 costs and
7000 taxes will
be paid. To ensure that the savings last until the expected lifetime,
a cheaper broker with only 1% percentage costs can be chosen to
minimize the costs. This will result in
3150 costs,
8120 taxes
and a leftover depot amount of
15700 at the expected lifetime
(still not enough though). In this case, the visualization enables
long-term planning and shows the impact of individual factors. At
best, it could lead to better future planning and decision-making.
For example, changing the bank and increase monthly investment
dynamically to minimize costs and allow for a buer respectively.
Discussion. The simulation has several advantages and a few dis-
advantages. First, it relies on a few recent years for forecasting. This
is hard to come by since ETFs have become popular only in recent
years. Next, the simulation is deliberately using a linear model in-
stead of an exponential model to prevent overestimation. Moreover,
the simulation is based on the current situation. Thus, it is based
on the current taxation laws and broker costs. Both will probably
change in the future, which would invalidate the simulation. The
visualization is good at displaying the negative factors costs, taxes
and ination, and the ratio between those negative factors and the
depot value. It covers both, the depot development and the cash ow
simultaneously. A drawback might be the unconventional layout
of the ination, which has been mentioned by several individuals
who got in touch with the visualization. The other negative values
are already discounted from the depot value, except the ination.
We decided to add a little hint by providing a visualization option
that discounts the ination from the total line. In its presented state,
the simulation and visualization are missing further explanations
for the parameters. For now, the user is expected to have basic
knowledge on ETFs and what factors are involved in a long-term
savings plan or retirement strategy. However, a certain learning and
exploration phase is expected. The user is expected to discover and
understand the parameters and their eect on the simulation and
visualization by means of iterative exploration of the problem and
solution space. A similar approach has been done by [
]. There,
a lot of dierent visualizations are oered, each, however, stands
for itself and does not allow for direct visual transfer between one
another. We think that the visualization is a meaningful resource
for individuals who intend to start ETF saving plans and want to
understand the impact of a certain strategies and gure out the
specics for their plan. Furthermore, the visualization could be ap-
plied to other use cases, e.g., the passive income for one’s children,
which would be handed over to them when they are grown up.
We implemented a prototypical simulation and visualization for an
ETF retirement plan strategy, which is based on the German cost and
taxation model. The prototype ( is build using Typescript,
React, Bootstrap, and D3, and source code (
vis) are publicly available. It and can be used for dierent use cases
next to the retirement planning such as planning a passive income
for one’s child. It might be able to ll the gap of personal visualiza-
tions which can “support people with limited visualization literacy
and analytics experience” [
] and, following, the supports under-
standing and actions to enact on one’s own retirement plans [
As a result, we extend available literature towards what is needed
to create eective, interactive visualization for the communication
and literacy of upcoming nancial products [13].
In order to evaluate the eectiveness of the visualization, a user
study with dierent groups regarding age, economic stand and
nancial literacy could be done. The visualization could be extended
to explain the underlying simulation. Moreover, the simulation
could be extended with further components, such as a simulation
of a nancial crash. That way, a user could interactively explore the
consequences on the retirement savings. We believe that the type
of visualization presented is a useful addition to the tools already
available and facilitates the planning and adjustment of long-term
savings goals and strategies.
Interactive Simulation and Visualization of Long-Term, ETF-based Investment Strategies VINCI ’21, September 6–8, 2021, Potsdam, Germany
This work is part of the “Software-DNA” project, which is funded
by the European Regional Development Fund (ERDF or EFRE in
German) and the State of Brandenburg (ILB).
Mahantesh C Angadi and Amogh P Kulkarni. 2015. Time Series Data Analysis
for Stock Market Prediction using Data Mining Techniques with R. International
Journal of Advanced Research in Computer Science 6, 6 (2015).
Adnan Balloch, Anamaria Nicolae, and Dennis Philip. 2014. Stock Market Literacy,
Trust, and Participation. Review of Finance 19, 5 (2014), 1925–1963. https:
BlackRock. [n.d.]. Der ETF-Sparplanrechner | iShares.
de/privatanleger/de/anlegen/sparen-mit- etfs/etf-sparplanrechner
Axel H Börsch-Supan, Michela Coppola, and Anette Reil-Held. 2012. Riester pen-
sions in Germany: design, dynamics, targetting success and crowding-in. Technical
Report. National Bureau of Economic Research.
Jiajia Cheng, Hu nan Deng, Guang Sun, Peng Guo, and J. Zhang. 2020. Application
of ARIMA Model in Financial Time Series in Stocks. In ICAIS.
Johann U. de Villiers and Elze-Mari Roux. 2019. Reframing the Retirement Saving
Challenge: Getting to a Sustainable Lifestyle Level. Journal of Financial Counseling
and Planning 30, 2 (2019), 277–288.
extraETF. [n.d.]. ETF-Sparplanrechner | ETF-Sparpläne einfach berechnen. https:
Aaron Fry, Jennifer Wilson, and Carol Overby. 2013. Teaching the design of
narrative visualization for nancial literacy. Art, Design & Communication in
Higher Education 12, 2 (2013), 159–177.
Kirti Goyal and Satish Kumar. 2021. Financial literacy: A systematic review
and bibliometric analysis. International Journal of Consumer Studies 45, 1 (2021),
Dandan Huang, Melanie Tory, Bon Adriel Aseniero, Lyn Bartram, Scott Bateman,
Sheelagh Carpendale, Anthony Tang, and Robert Woodbury. 2015. Personal
Visualization and Personal Visual Analytics. IEEE Transactions on Visualization
and Computer Graphics 21, 3 (2015), 420–433.
Gerd Kommer. 2018. Souverän investieren mit Indexfonds und ETFs: Wie Privatan-
leger das Spiel gegen die Finanzbranche gewinnen. Campus Verlag.
Markus Leippold, Lujing Su, and Alexandre Ziegler. 2016. How index futures
and ETFs aect stock return correlations. SSRN (2016).
Annamaria Lusardi, Anya Samek, Arie Kapteyn, Lewis Glinert, Angela Hung,
and Aileen Heinberg. 2017. Visual tools and narratives: new ways to improve
nancial literacy. Journal of Pension Economics and Finance 16, 3 (2017), 297—-323.
A. Mohamedashik, K. Senthamaraikannan, and D. Balamurugan. 2019. Statistical
Techniques For Stock Price Prediction In India.
Lucas Nunno. 2014. Stock Market Price Prediction Using Linear and Polynomial
Regression Models.
Anshul Vikram Pandey, Anjali Manivannan, Oded Nov, Margaret Satterthwaite,
and Enrico Bertini. 2014. The Persuasive Power of Data Visualization. IEEE
Transactions on Visualization and Computer Graphics 20, 12 (2014), 2211–2220.
Stephen Rudolph, Anya Savikhin, and David S. Ebert. 2009. FinVis: Applied
visual analytics for personal nancial planning. In 2009 IEEE Symposium on
Visual Analytics Science and Technology. 195–202.
Tyler, Portfolio Charts. [n.d.]. Methodology – Portfolio Charts. https://
Norman Hans-Peter Ubber. [n.d.]. Exchange Traded Fund Savings Plans as Old-
age Provision Tool in Germany. In Brno, November 28, 2019 European Scientifc
Conference of Doctoral Students. 149.
Martin Wattenberg. 1999. Visualizing the Stock Market. In CHI ’99 Extended
Abstracts on Human Factors in Computing Systems (Pittsburgh, Pennsylvania)
(CHI EA ’99). Association for Computing Machinery, New York, NY, USA, 188–189.
Sonja Wendel, Meimei Dai, Benedict Dellaert, and Bas Donkers. 2016. Consumer
retirement planning over the life cycle. (2016).
Josh Wiener and Tabitha Doescher.2008. A Framework for Promoting Retirement
Savings. Journal of Consumer Aairs 42, 2 (2008), 137–164.
Edward N Wol. 2011. The transformation of the American pension system: Was it
benecial for workers? WE Upjohn Institute.
ResearchGate has not been able to resolve any citations for this publication.
Full-text available
Given the paucity of comprehensive summaries in the extant literature, this systematic review, coupled with bibliometric analysis, endeavors to take a meticulous approach intended at presenting quantitative and qualitative knowledge on the ever-emerging subject of financial literacy. The study comprises a review of 502 articles (Annexure 1) published in peer-reviewed journals during 2000-2019. Citation network, page-rank analysis, co-citation analysis, content analysis, and publication trends have been employed to identify influential work, delineate the intellectual structure of the field, and identify gaps. The most prominent journals, authors, countries, articles, and themes have been identified using bibliometric analysis, followed by a comprehensive analysis of the content of 107 papers in the identified clusters. The three major themes enumerated are - levels of financial literacy among distinct cohorts, the influence that financial literacy exerts on financial planning and behavior, and the impact of financial education. Additionally, content analysis of 175 papers has been conducted for the last four years’ articles that were not covered in the co-citation analysis. Emerging themes identified include financial capability, financial inclusion, gender gap, tax and insurance literacy, and digital financial education. A conceptual framework has been modeled portraying the complete picture, following which potential areas of research have been suggested. This study will help policy-makers, regulators and academic researchers know the nuts and bolts of financial literacy, and identify the relevant areas that need investigation.
Full-text available
Data visualization has been used extensively to inform users. However, little research has been done to examine the effects of data visualization in influencing users or in making a message more persuasive. In this study, we present experimental research to fill this gap and present an evidence-based analysis of persuasive visualization. We built on persuasion research from psychology and user interfaces literature in order to explore the persuasive effects of visualization. In this experimental study we define the circumstances under which data visualization can make a message more persuasive, propose hypotheses, and perform quantitative and qualitative analyses on studies conducted to test these hypotheses. We compare visual treatments with data presented through barcharts and linecharts on the one hand, treatments with data presented through tables on the other, and then evaluate their persuasiveness. The findings represent a first step in exploring the effectiveness of persuasive visualization.
Full-text available
Data surrounds each and every one of us in our daily lives, ranging from exercise logs, to archives of our interactions with others on social media, to online resources pertaining to our hobbies. There is enormous potential for us to use these data to understand ourselves better and make positive changes in our lives. Visualization (Vis) and visual analytics (VA) offer substantial opportunities to help individuals gain insights about themselves, their communities and their interests; however, designing tools to support data analysis in non-professional life brings a unique set of research and design challenges. We investigate the requirements and research directions required to take full advantage of Vis and VA in a personal context. We develop a taxonomy of design dimensions to provide a coherent vocabulary for discussing personal visualization and personal visual analytics. By identifying and exploring clusters in the design space, we discuss challenges and share perspectives on future research. This work brings together research that was previously scattered across disciplines. Our goal is to call research attention to this space and engage researchers to explore the enabling techniques and technology that will support people to better understand data relevant to their personal lives, interests, and needs.
Full-text available
This article studies the importance of stock market literacy and trust for stock ownership decisions. We find that these two distinct channels simultaneously explain not only the probability of participation, but, conditional on participation, also explain the share of investment in stocks. Once we account for stock market literacy, sociability is no longer significant for participation; what matters is literacy rather than sociability. Further, we observe that economic shocks and future expectations are key behavioral characteristics that explain a household’s decision to invest in stocks. However, upon participation, a larger set of behavioral characteristics explains the level of stock investment.
Full-text available
The authors provide a scholarly definition for metaphor-rich, story-driven ‘narrative visualization’. They argue that metaphors create a rich and emotionally resonant set of associations that frame the narrative and effectively support ‘System 1’ (or intuition-based) thinking and decision-making that Daniel Kahneman and others have identified as the primary drivers of financial behaviour. The authors then apply these observations to a case study in which they analyse student work on a financial literacy design project. They discuss best practices for teaching narrative visualization and argue for its relevance in a contemporary design education – especially its capacity to represent and reflectively explore complex financial and other concepts.
An increasing number of individuals will be unable to retire comfortably amidst an international retirement savings crisis. Research suggests that behavioral factors contribute to inadequate retirement savings. We present a procedure that reframes the retirement savings decision, aimed at alleviating some of the negative effects of the behavioral factors. This procedure shifts the focus from the required wealth at retirement (the future) to the lifestyle an individual can afford to maintain now (the present). A sustainable lifestyle level (SLL) approach is expressed mathematically and illustrated with practical examples. The SLL approach offers a practical tool for retirement planning professionals to present recommendations that are simple and easy to understand for individuals faced with complex retirement planning decisions.
We examine both theoretically and empirically whether increased trading activity in index futures and exchange traded funds (ETFs) is associated with higher equity return correlations. Our model predicts that demand shocks to ETFs and futures lead to stronger price comovement for index stocks and non-index stocks. Moreover, demand shocks to ETFs have a higher impact on stock return correlations than shocks to futures. We confirm the model predictions by studying the correlation of U.S. stocks after the inception of S&P 500 futures and ETFs. Furthermore, our empirical results suggest that the return comovement induced by index trading is excessive.
We developed and experimentally evaluated four novel educational programs delivered online: an informational brochure, a visual interactive tool, a written narrative, and a video narrative. The programs were designed to inform people about risk diversification, an essential concept for financial decision-making. The effectiveness of these programs was evaluated using the American Life Panel. Participants were exposed to one of the programs, and then asked to answer questions measuring financial literacy – in particular, risk literacy – and self-efficacy. All of the programs were found to be effective at increasing self-efficacy, and several improved financial literacy, providing new evidence for the value of programs designed to improve financial decision-making.
Riester pensions are voluntary, but heavily subsidized private pension schemes in Germany. They were designed as a matching defined contribution scheme to fill the emerging “pension gap” that is being generated by the gradually declining generosity of the public pay-as-you-go pensions in response to population aging. This paper investigates how the uptake of the recently introduced “Riester pensions” depends on the state-provided saving incentives and how well the targeting to families and low-income households has worked in practice. It documents the costs of the scheme, and collects circumstantial evidence on displacement effects between saving for old-age provision and other purposes.After a slow start and several design changes, Riester pension plans took off very quickly. While saving incentives were effective in reaching parents, they were somewhat less successful in attracting low-income earners, although Riester pensions exhibit a more equal pattern by income than occupational pensions and unsubsidized private pension plans. Riester pension savings totaled €9.4bn in 2010 with an associated cost of €3.5bn. One average one Euro of subsidies is thus associated with 2 Euros of households’ own Riester saving.There is no evidence that Riester pensions have crowded out other saving. While households who plan to purchase housing and who attach high importance to a bequest motive are less likely to have a Riester pension, several regression results show that occupational pensions and other forms of private pensions act as complements rather than as substitutes. Aggregate national saving has increased since the introduction of Riester pensions.Institutional subscribers to the NBER working paper series, and residents of developing countries may download this paper without additional charge at
We describe a new 2-dimensional visualization algorithm capable of presenting detailed information on hundreds of items while emphasizing overall patterns in the data. This display method, which builds on Shneiderman's treemap technique, makes use of both hierarchy and similarity information. We have implemented this display in the SmartMoney Map of the Market, a web page that reports current data on over 500 publicly traded companies.