Package 'agriME'

Agricultural Marketing Efficiency and Price Spread Analysis

Description

Tools for reproducible price-spread, producer-share, marketing-cost, intermediary-margin, and marketing-efficiency analysis using channel totals or stage-level market-chain accounts.

Details

The principal workflows are marketing_metrics() for channel totals, analyse_channels() for stage-level accounts, and bootstrap_marketing_metrics() for repeated observations. All monetary inputs used in one comparison should refer to an equivalent commodity quality, form, place, time, and quantity unit.

Author(s)

Chiranjit Mazumder, Bikramjeet Ghose, and Pramit Pandit.

References

Acharya, S. S. and Agarwal, N. L. (2021). Agricultural Marketing in India, 7th edition. Oxford and IBH. ISBN 9789389688061.

Shepherd, A. W. (2007). A Guide to Marketing Costs and How to Calculate Them. Food and Agriculture Organization of the United Nations. https://www.fao.org/4/u8770e/u8770e00.htm

See Also

marketing_metrics, analyse_channels, bootstrap_marketing_metrics

---

Standardise, Validate, and Analyse Stage-Level Marketing Channels

Description

Creates an explicit market-channel data contract, detects inconsistent accounts, and calculates actor-level margins plus channel-level price-spread and efficiency statistics.

Usage

as_market_channel(
  data,
  channel = "channel",
  stage = "stage",
  actor = "actor",
  actor_type = "actor_type",
  purchase_price = "purchase_price",
  sale_price = "sale_price",
  marketing_cost = "marketing_cost",
  quantity = NULL,
  loss_percent = NULL,
  strict = FALSE
)

validate_market_channel(data, tolerance = 1e-08)

analyse_channel(data, tolerance = 1e-08, strict = FALSE)

analyse_channels(data, tolerance = 1e-08, strict = FALSE)

Arguments

data	A data frame containing one row per market actor and channel stage, or a standard market_channel object.
channel, stage, actor, actor_type, purchase_price, sale_price, marketing_cost	Column names in data.
quantity	Optional quantity column name.
loss_percent	Optional physical-loss percentage column name.
strict	If TRUE, broken inter-stage price links stop the operation.
tolerance	Non-negative numerical tolerance for matching prices and accounting identities.

Details

Each channel must contain exactly one producer at the first stage. All later rows must have actor_type = "intermediary". The sale price at one stage should equal the purchase price at the next stage when prices refer to the same product form and quantity. A direct channel may contain only the producer row; the producer sale price is then also the consumer price.

The producer net price is the producer sale price minus the producer marketing cost. Intermediary net margin is sale price minus purchase price minus actor marketing cost.

Value

as_market_channel() returns a data frame of class market_channel. validate_market_channel() returns a diagnostics data frame. The analysis functions return an agriME_analysis object with summary, actors, and diagnostics components.

Examples

data(tomato_channels)
channel_data <- as_market_channel(tomato_channels)
validate_market_channel(channel_data)

fit <- analyse_channels(channel_data)
fit
summary(fit)
fit$actors

---

Bootstrap Confidence Intervals for Marketing Metrics

Description

Resamples repeated observations independently within each channel and calculates nonparametric percentile intervals for price-spread, producer-share, and marketing-efficiency indicators.

Usage

bootstrap_marketing_metrics(
  data,
  producer_price = "producer_price",
  consumer_price = "consumer_price",
  marketing_cost = "marketing_cost",
  marketing_margin = NULL,
  channel = NULL,
  weight = NULL,
  R = 999L,
  conf = 0.95,
  seed = NULL,
  na.rm = FALSE,
  shepherd_variant = c("ratio", "net_ratio")
)

Arguments

data	Observation-level data frame.
producer_price, consumer_price, marketing_cost	Column names or numeric vectors.
marketing_margin	Optional column name or numeric vector. If NULL, margin is derived from the accounting identity for every row.
channel	Optional grouping column name or vector. The default pools all observations.
weight	Optional non-negative weight column or vector.
R	Integer number of bootstrap replicates, at least 20.
conf	Confidence level strictly between zero and one.
seed	Optional finite random seed. The previous random-number state is restored when the function exits.
na.rm	Remove incomplete observations.
shepherd_variant	Shepherd definition passed to marketing_metrics().

Details

Point estimates and bootstrap replicates are calculated from channel means. If weights are supplied, weighted means are used within each resample. Each channel must contain at least two complete observations. Percentile intervals quantify sampling variability under the empirical resampling scheme; they do not correct survey-design bias or establish causal channel effects.

Value

An agriME_bootstrap object with a long summary table, a wide replicates table, the matched call, and the Shepherd definition.

Examples

data(market_observations)
boot <- bootstrap_marketing_metrics(
  market_observations,
  marketing_margin = "marketing_margin",
  channel = "channel",
  weight = "volume_qtl",
  R = 49,
  seed = 2026
)
boot

---

Decompose the Consumer Price

Description

Decomposes each unit of consumer expenditure into net producer price, marketing cost, net intermediary margin, and any accounting gap.

Usage

consumer_rupee(
  x = NULL,
  producer_price = NULL,
  consumer_price = NULL,
  marketing_cost = NULL,
  marketing_margin = NULL,
  channel = NULL
)

Arguments

x	An agriME_analysis object, or a numeric producer price.
producer_price	Alternative named net producer-price input when x is not supplied.
consumer_price	Consumer price.
marketing_cost	Total marketing cost.
marketing_margin	Total net intermediary margin; derived if omitted.
channel	Optional channel labels for direct numeric input.

Value

A long data frame with channel, component, monetary value, and percentage of consumer price.

Examples

consumer_rupee(
  producer_price = 1900,
  consumer_price = 3150,
  marketing_cost = 510,
  marketing_margin = 740,
  channel = "Wholesale-retail"
)

data(tomato_channels)
consumer_rupee(analyse_channels(tomato_channels))

---

Marketing Margin after Physical Product Loss

Description

Adjusts acquisition cost and trading margin for handling, transport, storage, processing, or spoilage losses.

Usage

loss_adjusted_margin(
  purchase_price,
  sale_price,
  quantity_purchased,
  quantity_sold,
  marketing_cost = 0,
  cost_basis = c("lot", "purchased_unit", "sold_unit")
)

Arguments

purchase_price	Price per purchased unit.
sale_price	Price per sold unit.
quantity_purchased	Quantity acquired before loss.
quantity_sold	Quantity sold after loss.
marketing_cost	Marketing cost stated according to cost_basis.
cost_basis	Whether marketing cost is for the complete lot, each purchased unit, or each sold unit.

Details

The effective purchase cost per sold unit is total purchase value divided by quantity sold. Gross margin is sales value minus purchase value; net margin also deducts total marketing cost. Quantity sold cannot exceed quantity purchased.

Value

A data frame containing quantity loss, loss percentage, purchase and sales values, total marketing cost, effective unit cost, and gross and net margins.

Examples

loss_adjusted_margin(
  purchase_price = 20,
  sale_price = 28,
  quantity_purchased = 100,
  quantity_sold = 92,
  marketing_cost = 180,
  cost_basis = "lot"
)

---

Repeated Illustrative Agricultural Market Observations

Description

Synthetic weekly channel-level observations designed for bootstrap examples. The accounting fields obey the price identity. These data are not empirical survey results.

Usage

market_observations

Format

A data frame with 72 rows and 10 variables:

date

Weekly observation date.

market

Illustrative market name.

commodity

Commodity name.

unit

Price unit.

channel

Marketing channel.

producer_price

Net producer price in Indian rupees per quintal.

consumer_price

Final consumer price in Indian rupees per quintal.

marketing_cost

Total channel marketing cost.

marketing_margin

Total net intermediary margin.

volume_qtl

Illustrative transaction volume in quintals.

Source

Synthetic data created by the package authors.

Examples

data(market_observations)
head(market_observations)

---

Agricultural Marketing Efficiency and Complete Channel Metrics

Description

Calculates alternative marketing-efficiency indices or a comprehensive table of price-spread and efficiency results.

Usage

marketing_efficiency(
  producer_price,
  consumer_price,
  marketing_cost,
  marketing_margin = NULL,
  method = c("acharya", "shepherd", "conventional"),
  shepherd_variant = c("ratio", "net_ratio")
)

marketing_metrics(
  producer_price,
  consumer_price,
  marketing_cost,
  marketing_margin = NULL,
  channel = NULL,
  shepherd_variant = c("ratio", "net_ratio")
)

Arguments

producer_price	Net price received by the producer per common unit.
consumer_price	Price paid by the final consumer per the same unit.
marketing_cost	Total marketing cost per the same unit.
marketing_margin	Total net intermediary margin. If NULL, it is derived as consumer price minus producer price minus marketing cost.
method	One of "acharya", "shepherd", or "conventional".
shepherd_variant	"ratio" uses consumer price divided by marketing cost. "net_ratio" subtracts one from that ratio.
channel	Optional channel labels.

Details

The package implements the following accounting definitions:

$ME_A = P_f/(MC + MM)$

for Acharya efficiency,

$ME_S = P_c/MC$

for the Shepherd ratio, and

$ME_C = (P_c - P_f)/MC$

for conventional efficiency. Here $P_f$ is net producer price, $P_c$ is consumer price, $MC$ is total marketing cost, and $MM$ is total net intermediary margin. The optional net-ratio Shepherd variant is $P_c/MC - 1$.

When the supplied accounts satisfy $P_c=P_f+MC+MM$, Acharya efficiency is also equal to $P_c/(MC+MM)-1$. Any departure is retained in the accounting_gap column.

Value

marketing_efficiency() returns a numeric vector. marketing_metrics() returns a data frame containing all price, spread, share, accounting, and efficiency measures.

References

Acharya, S. S. and Agarwal, N. L. (2021). Agricultural Marketing in India, 7th edition. Oxford and IBH. ISBN 9789389688061.

Examples

marketing_efficiency(1900, 3150, 510, 740, method = "acharya")
marketing_efficiency(1900, 3150, 510, method = "shepherd")

marketing_metrics(
  producer_price = c(2420, 1900),
  consumer_price = c(2600, 3150),
  marketing_cost = c(180, 510),
  marketing_margin = c(0, 740),
  channel = c("Direct", "Wholesale-retail")
)

---

Gross and Net Margin of a Market Intermediary

Description

Separates the intermediary's gross price margin from net margin after marketing cost.

Usage

marketing_margin(
  purchase_price,
  sale_price,
  marketing_cost = 0,
  consumer_price = NULL
)

Arguments

purchase_price	Purchase price per unit.
sale_price	Sale price per unit.
marketing_cost	Marketing cost incurred per unit.
consumer_price	Optional final consumer price used to calculate gross and net margin shares.

Details

Gross margin is sale price minus purchase price. Net margin is gross margin minus marketing cost. Markup is measured relative to purchase price; an undefined markup caused by a zero purchase price is returned as NA.

Value

A data frame containing purchase price, sale price, marketing cost, gross margin, net margin, markup percentage, and optional consumer-price shares.

Examples

marketing_margin(
  purchase_price = c(2050, 2450),
  sale_price = c(2450, 3150),
  marketing_cost = c(140, 220),
  consumer_price = 3150
)

---

Marketing Sensitivity and Efficiency-Target Analysis

Description

Evaluates a full factorial set of price, cost, and margin shocks, or solves an efficiency equation for the value required to attain a target.

Usage

marketing_sensitivity(
  producer_price,
  consumer_price,
  marketing_cost,
  marketing_margin = NULL,
  producer_change = c(-0.1, 0, 0.1),
  consumer_change = 0,
  cost_change = c(-0.1, 0, 0.1),
  margin_change = 0,
  shepherd_variant = c("ratio", "net_ratio")
)

efficiency_target(
  target,
  method = c("acharya", "shepherd", "conventional"),
  solve_for = c(
    "producer_price",
    "consumer_price",
    "marketing_cost",
    "marketing_margin"
  ),
  producer_price = NULL,
  consumer_price = NULL,
  marketing_cost = NULL,
  marketing_margin = NULL,
  shepherd_variant = c("ratio", "net_ratio")
)

Arguments

producer_price, consumer_price, marketing_cost, marketing_margin	Baseline or known accounting values. An omitted margin in sensitivity analysis is derived by identity.
producer_change, consumer_change, cost_change, margin_change	Vectors of proportional changes; 0.10 means a ten percent increase.
shepherd_variant	One of "ratio" or "net_ratio".
target	Desired positive efficiency ratio.
method	Efficiency equation to solve.
solve_for	Unknown component to calculate.

Details

Sensitivity scenarios change each accounting component independently. A non-zero accounting gap can therefore be informative: it shows that the chosen scenario is not a closed price identity unless the related components are also adjusted.

Acharya efficiency can solve for producer price, marketing cost, or marketing margin. Shepherd efficiency can solve for consumer price or marketing cost. Conventional efficiency can solve for producer price, consumer price, or marketing cost.

Value

marketing_sensitivity() returns an agriME_sensitivity data frame. efficiency_target() returns a one-row data frame containing the required value.

Examples

sens <- marketing_sensitivity(
  1900, 3150, 510, 740,
  producer_change = c(0, 0.05),
  cost_change = c(-0.1, 0, 0.1)
)
sens

efficiency_target(
  target = 2,
  method = "acharya",
  solve_for = "marketing_cost",
  producer_price = 1900,
  marketing_margin = 740
)

---

Methods for agriME Result Objects

Description

Print, summary, conversion, and base-graphics methods for package result objects.

Usage

## S3 method for class 'agriME_analysis'
plot(
  x,
  type = c("decomposition", "efficiency", "producer_share"),
  col = NULL,
  main = NULL,
  ...
)

## S3 method for class 'agriME_bootstrap'
plot(
  x,
  metric = "acharya_efficiency",
  col = "#2C7FB8",
  main = NULL,
  ...
)

## S3 method for class 'agriME_ranking'
plot(x, col = "#756BB1", main = NULL, ...)

## S3 method for class 'agriME_sensitivity'
plot(
  x,
  metric = "acharya_efficiency",
  col = NULL,
  main = NULL,
  ...
)

## S3 method for class 'agriME_analysis'
print(x, digits = 3L, ...)

## S3 method for class 'agriME_bootstrap'
print(x, digits = 3L, ...)

## S3 method for class 'agriME_ranking'
print(x, digits = 3L, ...)

## S3 method for class 'agriME_sensitivity'
print(x, digits = 3L, ...)

## S3 method for class 'agriME_analysis'
summary(object, ...)

## S3 method for class 'agriME_analysis'
as.data.frame(
  x,
  row.names = NULL,
  optional = FALSE,
  ...
)

Arguments

x, object	An object created by an agriME analysis function.
type	Analysis plot type.
metric	Metric name to plot.
col	Optional colour or colour vector.
main	Optional plot title.
digits	Number of decimal places printed.
row.names, optional	Arguments passed to as.data.frame().
...	Additional arguments passed to the underlying method.

Value

Plot methods return plotted values invisibly. Print methods return x invisibly. The summary and data-frame methods return the channel summary table.

Examples

data(tomato_channels)
fit <- analyse_channels(tomato_channels)
summary(fit)
as.data.frame(fit)

ranking <- rank_channels(fit)
ranking


plot(fit, type = "decomposition")
plot(ranking)

---

Price Spread, Producer Share, and Total Gross Marketing Margin

Description

Calculates three closely related farm-retail price indicators from equivalent producer and consumer prices.

Usage

price_spread(producer_price, consumer_price, percent = FALSE)

producers_share(producer_price, consumer_price)

total_gross_marketing_margin(
  producer_price,
  consumer_price,
  percent = FALSE
)

Arguments

producer_price	Net price received by the producer per common unit.
consumer_price	Price paid by the final consumer per the same unit.
percent	Logical; express the result as a percentage of the consumer price.

Details

The absolute price spread is

$PS = P_c - P_f,$

where $P_c$ is consumer price and $P_f$ is the net producer price. The percentage price spread is $100 PS/P_c$, while producer share is $100 P_f/P_c$. With a net producer price, the total gross marketing margin is numerically equal to the price spread.

Value

A numeric vector. Producer share is always returned as a percentage.

Examples

price_spread(1900, 3150)
price_spread(1900, 3150, percent = TRUE)
producers_share(1900, 3150)
total_gross_marketing_margin(1900, 3150)

---

Rank Alternative Agricultural Marketing Channels

Description

Uses transparent min-max normalisation and a weighted additive score to compare channels on multiple marketing outcomes.

Usage

rank_channels(
  x,
  indicators = c(
    "acharya_efficiency",
    "producer_share_percent",
    "price_spread_percent"
  ),
  directions = c("max", "max", "min"),
  weights = NULL,
  channel_col = "channel"
)

Arguments

x	An agriME_analysis object or channel-level data frame.
indicators	Names of numeric indicator columns.
directions	One "max" or "min" value per indicator.
weights	Optional non-negative weights. Equal weights are used by default and supplied weights are rescaled to sum to one.
channel_col	Channel identifier column when x is a data frame.

Details

A constant indicator receives a normalised score of one for every channel, so it does not create artificial differences. Rankings are descriptive and depend on the selected indicators, directions, and value judgements embodied in the weights.

Value

An agriME_ranking data frame containing original indicators, normalised indicators, composite score, and rank. Weights and directions are stored as attributes.

Examples

data(tomato_channels)
fit <- analyse_channels(tomato_channels)
rank_channels(fit)
rank_channels(fit, weights = c(0.5, 0.3, 0.2))

---

Illustrative Tomato Marketing Channels

Description

A synthetic and internally consistent stage-level dataset representing four tomato marketing channels. It is intended for teaching, examples, and software validation, not as empirical survey evidence.

Usage

tomato_channels

Format

A data frame with 10 rows and 9 variables:

channel

Channel label.

stage

Ordered stage number.

actor

Market participant.

actor_type

Producer or intermediary.

purchase_price

Purchase price in Indian rupees per quintal.

sale_price

Sale price in Indian rupees per quintal.

marketing_cost

Marketing cost in Indian rupees per quintal.

quantity_kg

Reference quantity in kilograms.

loss_percent

Illustrative physical loss percentage.

Source

Synthetic data created by the package authors.

Examples

data(tomato_channels)
tomato_channels

---